Simple arguments on consecutive power residues (Q877922): Difference between revisions

The author shows that the number of consecutive \(k\)-th power residues modulo \(m\) is bounded from above by \(m/n\), where \(n = n_k(m)\) is the least positive \(k\)-th power nonresidue modulo \(m\). In fact, assume that \(a+1, \ldots, a+R\) are \(k\)-th power residues mod \(m\), and let \(q = \lceil an/m \rceil\). Then \((a+i)n-mq\) is a \(k\)-th power nonresidue mod \(m\) for each \(i \in \{1, \dots, R\}\). But if \((a+i)n-mq \geq 0\) for some \(i\), then \((a+i)n-mq \geq n\). From \(an - qm < 0\) we now deduce inductively that \((a+R)n-mq < 0\), which implies the claim. This argument is then generalized to Euclidean complex quadratic number fields.