Eine gemeinsame Basis für die Theorie der Eulerschen Graphen und den Satz von Petersen

The main result states: Lete ₁,e ₂,e ₃ be three lines incident to the pointv (degv4) of the connected bridgeless graphG such thate ₁ ande ₃ belong to different blocks ifv is a cutpoint. Split the pointv in two ways: Lete ₁,e _j ,j=2, 3, be incident to a new pointv _1j and leave the remainder ofG unchanged, thus obtainingG _1j . Then at least one of the two graphsG ₁₂,G ₁₃ is connected and bridgeless. — A classical result ofFrink follows from this theorem which is the key to a simple proof of Petersen's theorem. Moreover, the above result can be used to prove practically all classical results on Eulerian graphs, including best upper and lower bounds for the number of Eulerian trails in a connected Eulerian graph. In the theory of Eulerian graphs, it can be viewed as the basis for good algorithms checking on several properties of this class of graphs.

---

---

Herrn Prof. Dr. H. Hornich zum 70. Geburtstag gewidmet     

Two questions in diophantine approximation

The first question is about a possible variation on Dirichlet's approximation theorem for linear forms x+y+z, wherex,y are restricted topositive integers. The second question, which turns out to be related to the first, is about approximation to elements in a power series fieldk((t ^–1)) by solutions of first order linear differential equationsx+y+z=0, wherex,y,z are polynomials int. Dedicated to Prof. Dr. E. Hlawka on the occasion of his 60th birthday     

The method of Iterated Defect-Correction and its application to two-point boundary value problems

Summary In Part II of this paper we present a rigorous analysis of the Iterated Defect-Correction — applied to two-point boundary value problems — which was introduced in Part I of this paper [1]. A complete proof of Theorem 5. 1 of [1] is given.     

Observation of B(s)(0) → D(s)(*)+ D(s)(*)- using e+ e- collisions and a determination of the B(s)-B(s) width difference ΔΓ(s)

We have made the first observation of B(s)(0)→D(s)(*)+ D(s)(*)- decays using 23.6 fb(-1) of data recorded by the Belle experiment running on the Υ(5S) resonance. The branching fractions are measured to be B(B(s)(0)→D(s)+ D(s)-)=(1.03(-0.32-0.25)(+0.39+0.26))%, B(B(s)(0)→D(s)(*±) D(s)(?))=(2.75(-0.71)(+0.83)±0.69)%, and B(B(s)(0)→D(s)*+ D(s)*-)=(3.08(-1.04-0.86)(+1.22+0.85))%; the sum is B[B(s)(0)→D(s)(*)+ D(s)(*)-]=(6.85(-1.30-1.80)(+1.53+1.79))%. Assuming B(s)(0)→D(s)(*)+ D(s)(*)- saturates decays to CP-even final states, the branching fraction determines the ratio ΔΓ(s)/cosφ, where ΔΓ(s) is the difference in widths between the two B(s)-B(s) mass eigenstates, and φ is a CP-violating weak phase. Taking CP violation to be negligibly small, we obtain ΔΓ(s)/Γ(s)=0.147(-0.030)(+0.036)(stat)(-0.041)(+0.042)(syst), where Γ(s) is the mean decay width.