On unconditionally closed sets and a certain conjecture of A. A. Markov (Q1117044)

\textit{A. A. Markov} asked [in Izv. Akad. Nauk. SSSR, Ser. Mat. 9, 3-64 (1945; Zbl 0061.042)] whether a group G admits a connected Hausdorff group topology as soon as for each subgroup \(H<G\) closed with respect to every Hausdorff group topology on G, the following holds: \(Card(G/H)\geq 2^{\aleph_ 0}?\) We show that the answer is `no'. The construction of an example involves some calculations of a medium complexity with wreath products of topological groups, as well as a portion of classical functional analysis. As a whole, our paper contributes to a rather decrepit though still alive area of topological algebra, namely, the `general topological algebra', which badly needs a new blood.