Canine circovirus (CanineCV) has a worldwide distribution in dogs and is occasionally detected in wild carnivorans, indicating its ability for cross-species transmission. However, fox circovirus, a subclade of CanineCV, has been identified exclusively in wild canids. We analyzed spleen samples from 159 grey wolves from the Northwest Territories, Canada, to investigate the molecular epidemiology of CanineCV and formulate hypotheses about virus ecology and evolution. Overall, 72 (45.3%) animals tested positive. Virus prevalence was similar between males and females, adults and juveniles, and across the investigated years and locations. CanineCV infection was not associated with a poor body condition. While the percentage of co-infections with canine parvoviruses, investigated in a previous study, was high (63/72, 87.5%), the rate of parvovirus infection in CanineCV-negative animals was significantly lower (48.3%, p < 0.001) and CanineCV infection was associated with a 7.5- and 2.4-fold increase in the risk of acquiring canine parvovirus 2 or canine bufavirus infections, respectively (odds ratios: 3.5-16.9 and 1.3-5.8). Although common risk factors cannot be ruled out, this suggests that CanineCV may facilitate parvoviral super-infections. Sequencing revealed high genetic diversity, further exacerbated by recombination. Of the 69 sequenced strains, 87.5% were fox circoviruses, five were related to a fox circovirus-like recombinant strain, and one belonged to a distant clade. In the phylogenetic analysis, the viruses were distributed according to sampling locations, with some viruses being geographically restricted. Different clades of viruses were identified in the same areas and over multiple years (2007-2019), indicating the co-existence of multiple endemic lineages. Phylogenetic analysis of all available complete fox circovirus genomes (32 from foxes and 15 from wolves from North America and Europe) demonstrated four lineages, each including sequences from this study. Within each lineage, strains segregated geographically and not by host. This implies that, although multiple lineages co-exist, viruses do not frequently move between locations. Finally, viruses from Europe and North America were mixed, indicating that the origin of the four lineages might predate the segregation of European and American wolf and fox populations. Given the high prevalence and diversity of fox circoviruses in wolves, these animals should be considered reservoir hosts for these viruses. Although we cannot exclude a lower susceptibility of dogs, the lack of fox circovirus in dogs could be due to environmental circumstances that prevented its spread to dogs. Given the high diversity and wild host specificity, we presume a long-lasting association between fox circovirus and canine hosts and hypothesize a higher likelihood of transmission from dogs to wild animals than vice versa. Further studies should investigate other sympatric wild species and additional locations to explore the reasons behind the marked difference in cross-species transmission dynamics among the various clades.