Gram's law in the theory of the Riemann zeta-function. I (Q2631314)

In this paper, the author continues his previous studies begun in [Izv. Math. 74, No. 4, 743--780 (2010; Zbl 1257.11080); translation from Izv. Ross. Akad. Nauk, Ser. Mat. 74, No. 4, 83--118 (2010)], which concern the distribution of the imaginary parts of zeros of the Riemann zeta-function \(\zeta(s)\) and the so-called Gram's law. Denoting by \(\vartheta(t)\) the increment of the argument of the function \(\pi^{-s/2}\Gamma(s/2)\) along the segment of endpoints \(s =1/2\) and \(s=1/2+it\), recall that, for any \(n>0\), the Gram point \(t_n\) is defined as the unique root of the equation \(\vartheta(t_n)=\pi\cdot(n-1)\). Now, as explained by the author, suppose \(0<\gamma_1<\gamma_2<\dots\leq \gamma_n \leq \gamma_{n+1}\leq\dots\) are positive ordinates of zeros of \(\zeta(s)\) numbered in ascending order (if several zeros have the same ordinates, one numerates them in an arbitrary way). Researchers of zeros of the Riemann zeta function observed that for values of \(n\) not very large, with the exception of a small part of the cases, the ordinate \(\gamma_n\) belongs to the interval \(G_n=(t_{n-1},t_n]\). Later, this phenomenon was called ``Gram's law'' or ``Gram's rule''. In this paper, some statements concerning this distribution of these imaginary parts of zeros of the Riemann zeta-function are established. Such assertions are connected with the ``Gram's law'' or ``Gram's rule'' as defined. In particular, the author gives a proof of several \textit{A. Selberg}'s formulae previously stated by him without proof in his former paper [in: 10. Skand. Mat. Kongr., København 1946, 187--200 (1947; Zbl 0030.05003)], and some other related contributions.