Residuacity and genus theory of forms (Q643536)

Let \(p \equiv 1 \bmod 4\) be a prime number, and write \(p = a^2 + b^2\) for integers \(a, b\) with \(a\) odd. Let \(c\) be a squarefree integer not divisible by \(p\), and let \(\varepsilon = T + U \sqrt{c}\) denote the fundamental unit of \(\mathbb Z[\sqrt{c}\,]\). If \((\frac cp) = +1\), and if the length of the continued fraction expansion of \(\sqrt{c}\) is odd (which is equivalent to \(T^2 - cU^2 = -1\)), then \(\varepsilon\) represents a quadratic residue modulo \(p\) (when \(\sqrt{c}\) is replaced by a solution of the congruence \(x^2 \equiv c \bmod p\)) if and only if \(2(aT+b)\) is a quadratic residue modulo \(p\). Since \((\frac ap) = +1\) and \(i \equiv \frac ab \bmod p\) is a solution of \(x^2 \equiv -1 \bmod p\), the residuacity of \(2(aT+b)\) is the same as that of \(2(T+i) \bmod p\). The authors' observation then follows from results of \textit{J. A. Brandler} [J. Number Theory 5, 271--286 (1973; Zbl 0272.12002)]. In this article, the authors offer a proof based on the arithmetic of binary quadratic forms. In the corrigendum several typos are corrected.