H. Grauert – författare

The present book grew out of introductory lectures on the theory offunctions of several variables. Its intent is to make the reader familiar, by the discussion of examples and special cases, with the most important branches and methods of this theory, among them, e.g., the problems of holomorphic continuation, the algebraic treatment of power series, sheaf and cohomology theory, and the real methods which stem from elliptic partial differential equations. In the first chapter we begin with the definition of holomorphic functions of several variables, their representation by the Cauchy integral, and their power series expansion on Reinhardt domains. It turns out that, in l:ontrast ~ 2 there exist domains G, G c en to the theory of a single variable, for n with G c G and G "# G such that each function holomorphic in G has a continuation on G. Domains G for which such a G does not exist are called domains of holomorphy. In Chapter 2 we give several characterizations of these domains of holomorphy (theorem of Cartan-Thullen, Levi''s problem). We finally construct the holomorphic hull H(G} for each domain G, that is the largest (not necessarily schlicht) domain over en into which each function holomorphic on G can be continued.

The present book grew out of introductory lectures on the theory offunctions of several variables. Its intent is to make the reader familiar, by the discussion of examples and special cases, with the most important branches and methods of this theory, among them, e.g., the problems of holomorphic continuation, the algebraic treatment of power series, sheaf and cohomology theory, and the real methods which stem from elliptic partial differential equations. In the first chapter we begin with the definition of holomorphic functions of several variables, their representation by the Cauchy integral, and their power series expansion on Reinhardt domains. It turns out that, in l:ontrast ~ 2 there exist domains G, G c en to the theory of a single variable, for n with G c G and G "# G such that each function holomorphic in G has a continuation on G. Domains G for which such a G does not exist are called domains of holomorphy. In Chapter 2 we give several characterizations of these domains of holomorphy (theorem of Cartan-Thullen, Levi's problem). We finally construct the holomorphic hull H(G} for each domain G, that is the largest (not necessarily schlicht) domain over en into which each function holomorphic on G can be continued.

1. The classical theorem of Mittag-Leffler was generalized to the case of several complex variables by Cousin in 1895. In its one variable version this says that, if one prescribes the principal parts of a merom orphic function on a domain in the complex plane e, then there exists a meromorphic function defined on that domain having exactly those principal parts. Cousin and subsequent authors could only prove the analogous theorem in several variables for certain types of domains (e. g. product domains where each factor is a domain in the complex plane). In fact it turned out that this problem can not be solved on an arbitrary domain in em, m ~ 2. The best known example for this is a "notched" bicylinder in 2 2 e . This is obtained by removing the set { (z , z ) E e 11 z I ~ !, I z 1 ~ !}, from 1 2 1 2 2 the unit bicylinder, ~ :={(z , z ) E e llz1 < 1, lz1 < 1}. This domain D has 1 2 1 2 the property that every function holomorphic on D continues to a function holo­ morphic on the entire bicylinder. Such a phenomenon never occurs in the theory of one complex variable. In fact, given a domain G c e, there exist functions holomorphic on G which are singular at every boundary point of G.

Of making many books there is no end; and much study is a weariness of the flesh. Eccl. 12.12. 1. In the beginning Riemann created the surfaces. The periods of integrals of abelian differentials on a compact surface of genus 9 immediately attach a g­ dimensional complex torus to X. If 9 ~ 2, the moduli space of X depends on 3g - 3 complex parameters. Thus problems in one complex variable lead, from the very beginning, to studies in several complex variables. Complex tori and moduli spaces are complex manifolds, i.e. Hausdorff spaces with local complex coordinates Z 1, ... , Zn; holomorphic functions are, locally, those functions which are holomorphic in these coordinates. th In the second half of the 19 century, classical algebraic geometry was born in Italy. The objects are sets of common zeros of polynomials. Such sets are of finite dimension, but may have singularities forming a closed subset of lower dimension; outside of the singular locus these zero sets are complex manifolds.

Of making many books there is no end; and much study is a weariness of the flesh. Eccl. 12.12. 1. In the beginning Riemann created the surfaces. The periods of integrals of abelian differentials on a compact surface of genus 9 immediately attach a g­ dimensional complex torus to X. If 9 ~ 2, the moduli space of X depends on 3g - 3 complex parameters. Thus problems in one complex variable lead, from the very beginning, to studies in several complex variables. Complex tori and moduli spaces are complex manifolds, i.e. Hausdorff spaces with local complex coordinates Z 1, ... , Zn; holomorphic functions are, locally, those functions which are holomorphic in these coordinates. th In the second half of the 19 century, classical algebraic geometry was born in Italy. The objects are sets of common zeros of polynomials. Such sets are of finite dimension, but may have singularities forming a closed subset of lower dimension; outside of the singular locus these zero sets are complex manifolds.

...Je mehr ich tiber die Principien der Functionentheorie nachdenke - und ich thue dies unablassig -, urn so fester wird meine Uberzeugung, dass diese auf dem Fundamente algebraischer Wahrheiten aufgebaut werden muss (WEIERSTRASS, Glaubensbekenntnis 1875, Math. Werke II, p. 235). 1. Sheaf Theory is a general tool for handling questions which involve local solutions and global patching. "La notion de faisceau s'introduit parce qu'il s'agit de passer de donnees 'locales' a l'etude de proprietes 'globales'" [CAR], p. 622. The methods of sheaf theory are algebraic. The notion of a sheaf was first introduced in 1946 by J. LERAY in a short note Eanneau d'homologie d'une representation, C. R. Acad. Sci. 222, 1366-68. Of course sheaves had occurred implicitly much earlier in mathematics. The "Monogene analytische Functionen", which K. WEIERSTRASS glued together from "Func- tionselemente durch analytische Fortsetzung", are simply the connected components of the sheaf of germs of holomorphic functions on a RIEMANN surface*'; and the "ideaux de domaines indetermines", basic in the work of K. OKA since 1948 (cf. [OKA], p. 84, 107), are just sheaves of ideals of germs of holomorphic functions.Highly original contributions to mathematics are usually not appreciated at first. Fortunately H. CARTAN immediately realized the great importance of LERAY'S new abstract concept of a sheaf. In the polycopied notes of his Semina ire at the E. N. S.

Der dritte und letzte Teil unserer Darstellung der Differential­ und Integralrechnung ist der Integrationstheorie im. Rn gewidmet. Er ist gedacht für Mathematik- und Physikstudenten des dritten und vierten Semesters. Zum Verständnis wird der Stoff von Band I und ein kleiner Teil des Stoffes von Band II vorausgesetzt. 1. Wir beginnen (in Kap. I) mit dem Lebesgueschen Integral im Rn. Anstelle des sehr speziellen euklidischen Maßes legen wir sogleich allgemeine Radonsche Maße zugrunde und beziehen auf diese Weise das Lebesgue-Stieltjes-Integral und die Integration über das Dirac­ sche b-Maß in unsere Theorie ein. Um den Umweg über das Rie­ mannsche Integral zu vermeiden, führen wir Radonsche Maße als (stetige) Linearformen auf einem Vektorraum von Treppenfunk­ tionen ein, also nicht, wie sonst üblich, auf dem Raum der stetigen Funktionen mit kompaktem Träger. Natürlich gelangt man auch hierdurch zum üblichen Integralbegriff. in § 2 ist wieder so gefaßt, daß sie Die Definition des Integrals sich unverändert auf allgemeinste Fälle überträgt, z. B. auf Funk­ tionen mit Werten in einem topologischen Vektorraum V. Selbst­ verständlich muß V ein lokal-konvexer Hausdorff-Raum sein, wenn man sinnvolle Ergebnisse erwarten will. Iq diesem Fall werden Funk­ tionsbereiche folgendermaßen erklärt: Es sei W c Rn X V eine offene Menge, so daß für jeden Punkt ~ERn der Durchschnitt ({d X V) n W nichtleer und konvex ist; ferner gebe es eine kompakte Menge KclR,11 mit (Rn - K) X {O} c W.

Der nun vorliegende zweite Teil der dreibändigen Darstellung der Differential- und Integralredmung ist der Differentialredlnung der Funktionen mehrerer reellen Veränderlichen und den gewöhnlidlen Differentialgleidlungen gewidmet. Er ist gedadlt etwa für Studenten im zweiten bis dritten Semester - dementsprechend wird vom Leser nur die Kenntnis des wesentlidlen Teils des Stoffs von Band I und dar­ über hinaus Bekanntschaft mit dem Begriff des Vektorraums erwartet. Die Autoren haben sidl wieder um einen strengen und systemati­ sdlen Aufbau der Theorie bemüht. Dabei waren sie bestrebt, unnötige Abstraktionen und Verallgemeinerungen zu vermeiden, sie haben jedodl gleidlzeitig versudlt, Definitionen und Methoden so zu bringen, daß sie sidl möglidlst unmittelbar auf allgemeinste Fälle übertragen lassen. Beispielsweise besagt die Definition der (totalen) Differenzierbarkeit (in anderen Worten): Eine reelle Funktion f, die in einer offenen Umgebung U eines Punktes X in einem Zahlenraum lRn erklärt ist, heißt in X o o differenzierbar, wenn es eine in X stetige Abbildung x -+ L1" von U in o n den dualen Raum Horn (lR , lR) gibt, so daß f(x) =f(x ) +L1" (x-x ) o o gilt. Diese Definition überträgt sidl auf den Fall, wo X Punkt eines o separierten topologisdlen Vektorraumes E ist und die Werte von f in einem ebensoldlen Vektorraum Fliegen.