The presentation analyses a set of humorous data, such as memes and short video clips, that include intertextual references to classic and popular films. The analysis is based on data collected from various social media in the Czech Republic at the time of the covid pandemic in 2020-21, and draws on the international database of covid-related humour (Kuipers 2021). The aim of the study is to identify how cinematic art is employed in humorous creations, describe two sets of relations: between the verbal and the visual elements, and between the original movies and the current situation indexed by the modern memes and video clips. The findings indicate that memes, clips and other instances of humour based on telecinematic art manifest a specific kind of intertextuality, where the prior humorous texts (comedies and humorous scenes from films) are ‘echoed’ (Hale 2018), i.e. the earlier texts are ‘revisited’ and the original humour is sustained or reactivated in the new creation that need not, by itself and without the recipients’ access to the shared cultural knowledge, appear as humorous or at all meaningful. One way of looking at this intertextuality is to see it as humorous recycling (humorous texts being turned into subsequent humorous texts) and intertextual play that, at the era of the digital media, involves frequent mash-ups and hybridity (Yus 2019), and creativity of forms (Vásquez 2019), while often having a critical stance (Tsakona and Chovanec 2020: 4). The study also pays attention to the shared cultural humorous repertoire, consisting of verbal quotes from films which are instantly recognizable by members of a particular cultural or linguistic community. The findings indicate that such cinematic one-liners are more commonly entextualized in humorous memes when referencing local (national) cinematic works. By contrast, where memetic humour draws on global (English-language) movies, it is based on the exploitation of visual forms rather than the intertextual recycling of and allusions to the original verbal component. This indicates different values of the verbal and the visual components in local as opposed to global contexts.