You seem to be referencing the Scottish Book, and specifically problem 153 (whether all Banach spaces have the approximation property), for the solution of which Mazur promised a live goose as prize. Said prize was later awarded to Per Enflo in 1972 for proving the claim false.
Added: A complete pdf version of the Scottish Book, translated in English and with some commentary by Ulam is available at
https://web.archive.org/web/20180428090844/http://kielich.amu.edu.pl/Stefan_Banach/pdf/ks-szkocka/ks-szkocka3ang.pdf
It also contains the original formulation of problem 153, which is not exactly the same as the one I had previously mentioned (though with a little effort I reckon that it's possible to reconstruct why they are considered equivalent).
Added (2): Following the comments by Gerald Edgar and Margaret Friedland, a piece of information that's worth mentioning here is that the Scottish Book was written while Banach was working in Lwów (today's L'viv, in Ukrainian). Wikipedia doesn't mention Banach ever being associated with Warsaw-based cultural instituitons: he was working in Lwów essentially all the time that international contingencies allowed it, and sometimes he was in Kraków.