Some comments on "Friend or Foe? Large canid remains from Pavlovian sites and their archaeozoological context", a paper by Wilczyński et al. (2020).

• There is an error in the caption of Fig. 1 in the paper of Wilczyński et al. (2020). • The canid hemimandibles in this figure are not from Dolní Vĕstonice II and Pavlov ISE. • These jaws were found at the Gravettian Předmostí site (Czech Republic) • We address further issues with the methodology and conclusion of this paper. Two canid morphotypes have been proposed for the middle Upper Palaeolithic site of Předmostí (Moravia, Czech Republic): Pleistocene wolf and Palaeolithic dog (Germonpré et al., 2012, 2015; Galeta et al., 2020). In Wilczyński et al. (2020), faunal assemblages from other Upper Palaeolithic Moravian sites were analyzed and those results used to project a similar, which we interpret as flawed, result for Předmostí: that only wild canids were present. We address issues with their methodology and argue that their conclusion, that dogs were not present at any of those studied sites, is based on cursory taphonomic analysis. Further, their projection of these results to Předmostí, an assemblage they did not study, is unsubstantiated, but does highlight the current dearth of taphonomic information that could aid the study of dog domestication. We also point out an important error in their Fig. 1, where the hemimandibles intended to show the natural variability in large canids from Dolní Vĕstonice II and Pavlov I SE, are in fact canid specimens from Předmostí. Since several of these canid specimens have been published by Germonpré et al. (2015), we find it imperative to address this error to ensure these data are correctly referenced in future works. [ABSTRACT FROM AUTHOR]