Repository navigation
Version 2.2.0 adds three chapters to Volume II and new material to both volumes. No result of version 2.1.0 is withdrawn. None of the additions is a step towards a proof of the Riemann Hypothesis, and the text says so where it matters.
New in Volume II:
- A new Chapter II.22, Iwaniec's approach and the level of the Thorin measure: the level alpha explicit formula, the Cauchy semigroup between levels, and a proved converse: positivity of the level alpha Weil functional on the test functions phi_0(1 ± cos ay), a at least 5, is equivalent to the quasi Riemann hypothesis at alpha, and one sign suffices. A computed counterexample shows that the full line functional is not a criterion.
- A new Chapter II.23, Ramanujan's Delta in Thorin form: the level alpha dictionary for L(s, Delta), the Deligne compound Poisson law, the Hecke cone of products of symmetric powers, the minimum of the compensated density at the origin, Putnam's theorem for Delta, the eta products and the flow of the zeros of Delta.
- A new Chapter II.24, What the Euler product and the functional equation force: a positive measure whose zeta function has the functional equation of zeta and a single pole is a multiple of the integers; rigidity of zeta under finite surgery with a positive Euler product; a comparison principle and residue rigidity; the positive minorant route shown to be the Riemann Hypothesis restated; and explicit obstructions (Beurling systems, the family F with an extra quadruple of zeros, a conductor 7 function of Davenport and Heilbronn type with a positive theta kernel). All results are proved or computed, and the chapter explains why each route stops short of RH.
- A section on the level step in Chapter II.24: positivity of the level alpha Weil functional does not pass to level alpha minus epsilon by any argument that uses only positivity, the functional equation and Hecke structure (an explicit function with a nonnegative prime measure passes the level alpha test and has a zero to the left of it, and no step is uniform over such a class); the step is equivalent to a Taylor radius bound, which gives a prime free restatement of RH; the Polya law of the obstructing functions is that of zeta plus an independent summand that is not of Lee and Yang type. All proved or computed, none a step towards RH.
- Further results of the authors' research: Li's coefficients written against the level alpha Thorin measure, with the kernel shown unique and positivity of the measure shown to give only the first coefficient; curves over finite fields at every level; formal zeta functions over F_5 with a nonnegative Euler product and zeros off the line; Gaussian damping of the Polya kernel, which does not give even QRH(1); the Toeplitz form at level one, whose least eigenvalue decays like N^-4; and the observation that the zero side alone allows half of all zeros off the line (Chapters II.12, II.22, II.24). All proved or computed.
- From three further unrefereed notes of Polson (November 2026), refereed before merging: for Ramanujan's Delta, an unconditional lower bound for the Thorin density at the origin, the Thorin measure as an argument with a corrected phase table, the Weil form of Delta with a Davenport and Heilbronn control that behaves the same way, eta^18 as a control, and the Yoshida analogue as an open problem (II.23); a heuristic measurement of the best point on a level set of the phase diagram (II.5); and the facts that Lambert ratios of the coefficients do not give real roots and that the Jensen test is far weaker than the Hankel test (II.4, II.20).
- The open problems chapter becomes II.25. The barrier method chapter is now a section of the phase diagram chapter (II.5) and the conditional certificates chapter a section of the quasi Riemann hypothesis chapter (II.21).
- Kolmogorov readings of the hitting time laws, interlacing of the zeros under the heat flow, and the Laplace order of the canonical law.
- The Pick test, the explained share and saturation for the Cauchy deconvolution, a defect formula, the Jackson family at the Iwaniec levels, and the visibility problem for the deconvolution ladder.
- Additions to the Cauchy deconvolution, Lagarias, ferromagnetic, amplitude and approximating clock chapters, and the certified count of zeros for L(s, chi_-4) that solves a problem of the truncation chapter.
New in Volume I:
- Smirnov thinnings with a window corollary and the Kolmogorov and Kuiper duals, a proof of infinite divisibility and self decomposability from the zeros, a uniform factor theorem at the basepoint, and a Feynman and Kac section in the Berry and Keating chapter (spectral lemma, trace formula, box and well theorems).
- Lee and Yang and Williams material: a field strip criterion and forward deformations in the Cauchy mixtures chapter, the Brownian modes of Williams' variable and Ginibre's inequality, the Gamma structure of Williams' variable shown not to be arithmetic, ferromagnetic realisations of the Polya law, and Polya's kernel as a spin coupled to an Ising spin.
- The status table of the path now sits in the route chapter, the free probability problems with their chapter, and the ball arithmetic computation for the heat trace in the heat trace chapter.
- Material on the reciprocal xi function, the basepoint one (the Delta column of the minimisation theorem, computed), the heat flow of Delta and the xi characteristic function, the Cauchy scale mixture and the deconvolution path, with thirteen new references checked against CrossRef.
The Davenport and Heilbronn function runs through these additions as a control: wherever a property holds for it as well, the text records that the property says nothing on RH. Both volumes were tightened sentence by sentence and overlapping sections were consolidated, so the volumes are shorter than in version 2.1.0 although they hold more. The catalog (1012 results) and the chapter summaries cover the new results.