On arguments of Gauss sums (Q5960274)

In the paper [\textit{H. L. Montgomery, R. C. Vaughan} and \textit{T. D. Wooley}, Proc. Camb. Philos. Soc. 118, No. 1, 21-33 (1995; Zbl 0835.11032)] it was assumed that if a positive integer \(k\) is fixed and \(p\) runs through the primes of type \(p\equiv 1\pmod k\), then \(\theta_1,\dots,\theta_m\) modulo 1 will be arbitrarily close to any given configuration an infinite number of times. The author proves that this is not so for \(k\) divided by 6. In other words, it is proved that arguments of Gaussian sums for exponents divisible by 6 do not form a dense set in common.