Over the past 2O years classical Hodge theory has undergone several generalizations of great interest in algebraic geometry. The papers in this volume reflect the recent developments in the areas of: mixed Hodge theory on the cohomology of singular and open varieties, on the rational homotopy of algebraic varieties, on the cohomology of a link, and on the vanishing cycles; L -realization of the intersection cohomology for the cases of singular varieties and smooth varieties with degenerating coefficients; applications of cubical hyperresolutions and of iterated integrals; asymptotic behavior of degenerating variations of Hodge structure; the geometric realization of maximal variations; and variations of mixed Hodge structure. N

The last book XIII of Euclid's Elements deals with the regular solids which therefore are sometimes considered as crown of classical geometry. More than two thousand years later around 1850 Schl fli extended the classification of regular solids to four and more dimensions. A few decades later, thanks to the invention of group and invariant theory the old three dimensional regular solid were involved in the development of new mathematical ideas: F. Klein (Lectures on the Icosa hedron and the Resolution of Equations of Degree Five, 1884) emphasized the relation of the regular solids to the finite rotation groups. He introduced complex coordinates and by means of invariant theory associated polynomial equations with these groups. These equations in turn describe isolated singularities of complex surfaces. The structure of the singularities is investigated by methods of commutative algebra, algebraic and complex analytic geometry, differential and algebraic topology. A paper by DuVal from 1934 (see the References), in which resolutions play an important rele, marked an early stage of these investigations. Around 1970 Klein's polynomials were again related to new mathematical ideas: V. I. Arnold established a hierarchy of critical points of functions in several variables according to growing com plexity. In this hierarchy Kleinls polynomials describe the "simple" critical points."

Let G be the group of rational points of a split connected reductive group over a nonarchimedean local field of residue characteristic p.LetI be a pro-p Iwahori subgroup of G and let R be a commutative quasi-Frobenius ring. If H = R[I\G/I] denotes the pro-p Iwahori- Hecke algebra of G over R we clarify the relation between the category of H-modules and the category of G-equivariant coefficient systems on the semisimple Bruhat-Tits building of G.IfR is a field of characteristic zero this yields alternative proofs of the exactness of the Schneider-Stuhler resolution and of the Zelevinski conjecture for smooth G-representations generated by their I-invariants. In general, it gives a description of the derived category of H-modules in terms of smooth G-representations and yields a functor to generalized (?, ?)-modules extending the constructions of Colmez, Schneider and Vigneras.

A characterization is given for the factorizations of almost simple groups with a solvable factor. It turns out that there are only several infinite families of these non-trivial factorizations, and an almost simple group with such a factorization cannot have socle exceptional Lie type or orthogonal of minus type. The characterization is then applied to study s-arc-transitive Cayley graphs of solvable groups, leading to a striking corollary that, except for cycles, a non-bipartite connected 3-arc-transitive Cayley graph of a finite solvable group is necessarily a normal cover of the Petersen graph or the Ho?man-Singleton graph.

The Galois theory of di?erence equations has witnessed a major evolution in the last two decades. In the particular case of q-di?erence equations, authors have introduced several di?erent Galois theories. In this memoir we consider an arithmetic approach to the Galois theory of q-di?erence equations and we use it to establish an arithmetical description of some of the Galois groups attached to q-di?erence systems.

Ce. volume. u.t fioJtme de. la Ve.M-ion de6~ve. du .te.x.tu du cOn6Vte.n.cV6 fia-i.tu a N-ice. au cowu., d' un CoUoque. qu-i f.>' Ij u.t .te.n.u du 23 au 27 Ju-in 1981. Comme. Ie. f.>ugge.ILe. Mn ~e., Ie. f.>uje..t, volon..t~e.me.n..t ILU.tfLe.-in..t, gfLav-i.ta-i.t gILof.>f.>o-modo au.toulL du upacu pILoje.c.t-i6f.> de. pe..t-i.te. d-ime.n- f.>-ion e..t du co UfLbe.f.> , ce.la f.>UfL un COfLpf.> aigebfL-ique.me.n..t elM. Ii f.>e.mble. que. ce. cho-ix deUbVte a-i.t Ue b-ie.n accu~ danf.> I' e.nf.>e.mble. pM lu pa.Jt:ttupan..tf.>. Le. Colloque. a IL(LMe.mble une. M-ixan..ta-in.e. de. f.>peu~.tu ve.n.uf.> de. d-i66eILe.n..tf.> paljf.> e..t nouf.> noUf.> f.>ommu e.n pa.Jt:ttcuUe.fL ILejo~ de. l'-im- pofL.tan..te. pa.Jt:ttc-ipatio n de. nof.> vo~-inf.> -i.taUe.nf.>. Le. pIL06Uf.>e.UfL VIEuVONNE n'aljaYt.t paf.> pu pa.Jt:ttupe.fL au colloque.