It is based on the Latin alphabet but includes certain letters (9) with diacritics: the acute accent – kreska: ⟨ć, ń, ó, ś, ź⟩; the overdot – kropka: ⟨ż⟩; the tail or ogonek – ⟨ą, ę⟩; and the stroke – ⟨ł⟩.
In loanwords they are usually replaced by ⟨k⟩,[a] ⟨w⟩, and ⟨ks⟩, respectively (as in nikab 'niqab', kwark 'quark', weranda 'veranda', sawanna 'savanna', ekstra 'extra', oksymoron 'oxymoron'), although some loanwords retain their original spelling (e.g., quiz, virga), and in a few cases both spellings are accepted (such as veto or weto, volt or wolt).
⟨é⟩ was historically used in native words prior to the 1891 spelling reform by the Academy of Learning, e.g., cztéry, papiéż (now cztery 'four', papież 'pope').
When giving the spelling of words, certain letters may be said in more emphatic ways to distinguish them from other identically pronounced characters.
The accented letters also have their own sections in dictionaries (words beginning with ⟨ć⟩ are not usually listed under ⟨c⟩).
The Polish alphabet is completely included in the Basic Multilingual Plane of Unicode.
A common test sentence containing all the Polish diacritic letters is the nonsensical Zażółć gęślą jaźń ('Yellow the ego with/of a gusle').