*Hyperelliptic jacobians without
CM and modular representations*

An explicit construction of hyperelliptic jacobians without nontrivial
endomorphisms will be presented. Some of our examples use the Steinberg
representation of certain finite Chevalley groups.

**September 21
Robert Vaughan (PSU)**

*On sigma-phi numbers*

Joint work with Kevin Weis will be described investigating the properties of a sequence of natural numbers similar to, but somewhat easier to study than, Carmichael numbers.

**September 28
Irwin Kra (Stonybrook): MASS Colloquium**

*Projective embeddings of some
modular curves*

For fixed values of the parameter $\tau$, theta functions of one variable $\theta[\chi](z,\tau)$ with characteristics $\chi$ can be used to study elliptic function theory. The corresponding theta constants $\theta[\chi](0,\tau)$ can be used to study function theory on surfaces represented by the action of subgroups of the modular group $\Gamma = \mbox{ PSL}(2,{\bf Z})$ on the upper half plane. Theta functions with integer (and, slightly more generally, half integer) characteristics have been studied classically. One is able to get more insight by consideringtheta functions with rational characteristics.

Let $\Gamma(k)$ be the level $k$ (assume for simplicity that $k$ is an odd integer) principal congruence subgroup acting on the upper half plane ${\bf H}^2$. Let $X_k$ be the compactification of ${\bf H}^2/\Gamma(k)$ obtained by filling in the punctures. We discuss projective embeddings of $X_k$ using modifications of theta constants with characteristics of the form $\left [\begin{array}{c} \frac{2l+1}{k} \\ 1 \end{array} \right ]$ and related automorphic forms constructed from derivatives of theta functions.

**October 5
Zifeng Yang (PSU)**

*Zeta Measures over Function
Fields*

Let A be integral domain of polynomials over a finite field of r elements. A is taken as an analogue of the ring of rational integers, with the "positive" elements in A being the monic polynomials. As an analogue of the zeta function over Z, the zeta function over A can be defined, and it is known that it can be analytically continued to a "complex plane" . In this talk, we will present some recent results in the measure theory of function fields and applications to the zeta function over A.

**October 12
Ed Formanek (PSU)**

*Two questions about free and
relatively free groups*

Questions of B. Plotkin and A. Myasnikov
are answered.

(1) Let S be the semigroup
of endomorphisms of a finitely generated free group. We show that
the only automorphisms of S are the obvious ones.

(2) Let F(r,c) be a free nilpotent
group of rank r and class c. We determine for which r and c there
are nontrivial elements of F(r,c) which are fixed by every automorphism
of F(r,c), and for which r and c the automorphism group of F(r,c) has a
nontrivial center.

**October 19
Winnie Li (PSU)**

*Coverings of curves with asymptotically
many rational points*

Infinite families of curves defined over a finite field $F_q$ of $q$ elements containing many rational points can be used to construct good long algebraic geometry codes. Denote by $N_q(g)$ the maximum possible number of rational points contained in a curve of genus $g$ defined over $F_q$. Ihara introduced the quantity $A(q)$ which is the limsup of the quotient $N_q(g)/g$ as the genus $g$ approaches infinity. Drinfeld and Vladut proved that $A(q) \le \sqrt q$. On the other hand, the work of Ihara, Tsfasman, Vladut and Zink showed that $A(q) = \sqrt q$ when $q$ is a square. Much less is known when $q$ is not a square. In this talk we'll review known optimal constructions and present new lower bounds for $A(q)$ with $q$ nonsquare, in joint work with Hiren Maharaj.

**October 24
Thomas Ernst (Uppsala) (SPECIAL LECTURE: 119 Boucke, 04:00pm)**

*A new expression for generalized
Vandermonde determinants*

The purpose of this talk is to present
a new expression for the generalized Vandermonde determinant (GVD) ${a}_{\lambda+\delta}$
and thus for the Schur function defined by $s_{\lambda} = \frac{a_{\lambda+\delta}}{a_{\delta}}$.
This expression contains a vector sum of elementary symmetric polynomials.
First the case GVD with two deleted rows is treated and then a formula
for an arbitrary GVD is proved by induction. It would be interesting
to try to extend this formula to arbitrary

symmetric polynomials. In
the process we also obtain an equivalence relation on the set of all GVD.

**October 26
John Dillon (NSA): Colloquium**

*Maps and Character Sums on Finite Fields*

Recent research on cyclic difference sets has uncovered some surprising connections with more traditional objects of study such as Gauss and Jacobi sums, Dickson (and exceptional) polynomials and quadratic forms as well as with some fundamental properties of BCH and 1st Order Reed-Muller codes. We shall discuss a number of these results and raise some questions for further research.

**November 2
Andrew Granville (U of Georgia): Colloquium**

*Distribution of
values of L(1,chi)*

In this talk joint work with Soundararajan will be described one consequence of which is a proof of a conjecture of Montgomery and Vaughan.

**November 9
Dale Brownawell (PSU)**

*Linear independence in positive
characteristic and divided derivatives*

Function fields in positive characteristic have long provided a rich analogue of number fields. In particular, Jing Yu has established a full analogue for t-modules of the Baker-Wuestholz Theorem on the linear independence of logarithms in commutative algebraic groups.

Unlike the classical case, the function field case involves a variable in the base field. So Laurent Denis had the idea of showing the linear independence of certain derivatives (with respect to this variable) of the logarithm of the most basic t-module, the Carlitz module. In joint work, we showed the linear independence of all divided derivatives of any non-zero logarithm of an algebraic value for any Drinfeld module.

Recently I was able to show the linear independence of all divided derivatives of all coordinates of any non-zero logarithm of a simple t-module (or a minimal extension thereof).

No background beyond a first-year course in algebra will be assumed.

**November 16
Sheeram Abhyankar (Purdue)**

*Symplectic groups and permutation
polynomials*

The linear group trinomial provides a mnemonic device for the recently discovered permutation polynomials of M\"uller-Cohen-Matthews, whereas the symplectic group equation generalizes them, thereby giving rise to strong genus zero coverings for characteristic two.

**November 30
Prof. Hourong Qin (Nanjing University and Columbia University)**

*Tame kernels and Tate kernels of quadratic number fields*

We give a brief introduction to some connections between algebraic K-theory
and number theory. Then we focus on the study of the tame kernels and Tate
kernels of quadratic number fields. The results will be applied to the
solvability of the Pell's equation.

**December 7
Andy Pollington (BYU): Colloquium**

*Inhomogeneous Diophantine approximation
and a conjecture of Barnes and Swinnerton-Dyer concerning indefinite, binary
quadratic forms*

The Barnes Swinnerton-Dyer states
that the inhomogeneous minimum of an indefinite, rational, binary, quadratic
form is always rational and isolated in the inhomogeneous spectrum of the
form. We confirm this conjecture in certain cases. The evaluation
of such minima is closely related to the class number of real quadratic
fields.

**December 12
Daniel Bertrand (Paris VI, visiting IAS)**
**
Special lecture, 1:25-2:15, 116 McAllister.**

*Unipotent representations of classical and differential Galois
groups***
**

Unipotent representations of the
absolute Galois group of a number field naturally appear in the study of
the $l$-adic realizations of a one-motive $M$. In his work on deficient
one-motives, K. Ribet showed that the image of these representations become
small when $M$ presents an `antisymmetric' autoduality. Similarly,
reducible differential operators $L= L_t...L_1$ yield unipotent representations
of differential Galois groups. We shall explain how the size of their image
reflects the `complexity' of the product, and (when $t=3$) that it becomes
small iff $L$ presents a 'symmetric' autoduality. We shall further explain
how Grothendieck's notion of blended extensions provides a uniform explanation
of these phenomena.