The paper deals with lower bounds for the remainder term in asymptotics for a certain class of arithmetic functions. Typically, these are generated by a Dirichlet series of the form ζ 2(s)ζ(2s−1)ζ M(2s)H(s), where M is an arbitrary integer and H(s) has an Euler product which converges absolutely for R s > σ0, with some fixed σ0 < 1/2.

Let $\mathbb{N}$ be the set of positive integers and let $s\in \mathbb{N}$. We denote by $d^s$ the arithmetic function given by $d^s\left(n\right)={\left(d\left(n\right)\right)}^s$, where $d\left(n\right)$ is the number of positive divisors of $n$. Moreover, for every $\ell ,m\in \mathbb{N}$ we denote by ${\delta }^{s,\ell ,m}\left(n\right)$ the sequence $$\underset{m\text{-times}}{\underbrace{d^s(d^s(...d^s(d^s\left(n\right)+\ell )+\ell ...)+\ell )}}=\left\{\begin{array}{cc}{d}^s\left(n\right)\hfill & \text{for}m=1,\hfill \\ d^s(d^s\left(n\right)+\ell )\hfill & \text{for}m=2,\hfill \\ d^s(d^s(d^s\left(n\right)+\ell )+\ell )\hfill & \text{for}m=3,\hfill \\ ⋮\hfill \end{array}\right.$$ We present classical and nonclassical notes on the sequence ${\left({\delta }^{s,\ell ,m}\left(n\right)\right)}_{m\ge 1}$, where $\ell$, $n$, $s$ are understood as parameters.

Given an integer $n\ge 3$, let $u_1,...,u_n$ be pairwise coprime integers $\ge 2$, $𝒟$ a family of nonempty proper subsets of $\{1,...,n\}$ with “enough” elements, and $\epsilon$ a function $𝒟\to \{\pm 1\}$. Does there exist at least one prime $q$ such that $q$ divides ${\prod }_{i\in I}{u}_i-\epsilon \left(I\right)$ for some $I\in 𝒟$, but it does not divide $u_1\cdots u_n$? We answer this question in the positive when the $u_i$ are prime powers and $\epsilon$ and $𝒟$ are subjected to certain restrictions.We use the result to prove that, if ${\epsilon }_0\in \{\pm 1\}$ and $A$ is a set of three or more primes that contains all prime divisors of any number of the form ${\prod }_{p\in B}p-{\epsilon }_0$ for...

What should be assumed about the integral polynomials $f₁\left(x\right),...,f_k\left(x\right)$ in order that the solvability of the congruence $f₁\left(x\right)f₂\left(x\right)\cdots f_k\left(x\right)\equiv 0\left(modp\right)$ for sufficiently large primes p implies the solvability of the equation $f₁\left(x\right)f₂\left(x\right)\cdots f_k\left(x\right)=0$ in integers x? We provide some explicit characterizations for the cases when $f_j\left(x\right)$ are binomials or have cyclic splitting fields.

The technique developed by A. Walfisz in order to prove (in 1962) the estimate $H\left(x\right)\ll {(logx)}^{2/3}{(loglogx)}^{4/3}$ for the error term $H\left(x\right)={\sum }_{n\le x}\frac{\phi \left(n\right)}{n}-\frac{6}{{\pi }^2}x$ related to the Euler function is extended. Moreover, the argument is simplified by exploiting works of A.I. Saltykov and of A.A. Karatsuba. It is noted in passing that the proof proposed by Saltykov in 1960 of $H\left(x\right)\ll {(logx)}^{2/3}{(loglogx)}^{1+ϵ}$ is erroneous and once corrected “only” yields Walfisz’ result. The generalizations obtained can be applied to error terms related to various classical - and less classical - arithmetical...