Primes between consecutive squares and the Lindelöf hypothesis (Q377820)

In a previous paper [Funct.\ Approximatio, Comment.\ Math. 45, No. 2, 255--263 (2011; Zbl 1260.11058)] the author proved that, under the Lindelöf hypothesis that \(\zeta(\tfrac{1}{2}+it) \ll_\varepsilon | t| ^\varepsilon\), every interval \([n^2,\, (n+1)^2]\) lying in \([1, N]\) contains the expected number of primes, with at most \(O(N^\varepsilon)\) exceptions. This is a step towards the classical conjecture: ``between every two squares of natural numbers there is a prime''. In the present paper the author reproves the above result, assuming two different conjectures, each of which is implied by the Lindelöf hypothesis. One of these hypotheses involves an upper bound for the integral of the \(2k\)-th power of an expression involving the Chebyshev function \(\psi(x) = \sum_{p^a\leq x}\log p\). The other hypothesis deals with an upper bound for the zero-counting function \(N^{(k)}(\sigma, T)\). This denotes the ordered set \(\rho_j = \beta_j+i\gamma_j\;(1\leq j \leq 2k)\) of complex zeta-zeros for which \(\beta_j \geq \sigma, | \gamma_j| \leq T\), and \[ | \gamma_1 + \cdots + \gamma_k - \gamma_{k+1} - \cdots - \gamma_{2k}| \leq 1. \] The key ingredient in the proofs is a lemma due (op.\ cit.) to the author.