Evolutionary theory predicts that the species of an evolutionarily successful taxon would not overlap in spatial distribution. To test the prediction, we document our research on the spatial associations of mustelids, an evolutionarily successful group of order Carnivore, using infrared camera trap data on species distribution collected from the national nature reserves (NNRs) of Liancheng, Wolong, Tangjiahe and Heizhugou in China in 2017–2021. Data showed seven mustelid species occurring in the study area, including Arctonyx collaris , Meles leucurus , Martes foina , Martes flavigula , Mustela altaica , Mustela nivalis and Mustela sibirica . Following Ricklef’s definition of biological community, we identified five networks of species associations. The mustelids occurred in the networks. Species from the same genus, such as M. foina and M. flavigula , stayed in different networks to avoid competition owing to similar feeding habits or habitat preferences. Species with different feeding habits or habitat preferences either occurred in different networks, such as M. altaica and M. flavigula , or coexisted in the same networks but avoided direct spatial associations, such as M. foina and A. collaris . Asymmetrical associations were found between different genera, such as M. foina and M. altaica , or between different subfamilies, such as M. flavigula and A. collaris . These associations may be attributed to interspecific killing or seed dispersal. However, these associations accounted for only a small proportion and would not impact the species diversity of Mustelidae. It is thus concluded that the prediction is supported by our research findings and that spatial avoidance may be the biogeographic strategy of maintaining the species diversity of the family. We also found that the well protection of the mustelids may benefit the overall biodiversity conservation in Heizhugou, an NNR that has experienced severe deforestation.

In order to understand the community environment of giant panda and red panda, so as to increase the understanding of their ecological evolution and facilitate conservation and management. In our study, spatial data monitored by infrared camera from 2017 to 2018 in Wolong, China was used to analyze the spatial association network of terrestrial animals. The results show that a total of 35 terrestrial species is recorded, of which 20 species formed a spatial network including giant panda and red panda. In the network, giant panda and red panda are directly related, and 9 other species have direct spatial associations with them. Further analysis show that: (1) Giant panda and red panda already ate bamboo at the early stage of community evolution. Bamboo eating helps them blend into the community and coexist. (2) Giant panda had moderate niche separation with most of the species that have directly spatial associations, the same to red panda. There is a commensalism relationship between giant panda and red panda: red panda may create suitable habitat for giant panda to some extent, which is beneficial to the survival of giant panda. (3) In the existing community, giant panda has no natural enemies, and has a mutually beneficial relationship with golden snub-nosed monkey. (4) In the community, red panda has predator: the yellow-throated marten. Yellow-throated marten not only preys on red panda, but also may have evolved a unknown counter-anti-predatory strategy to attract red panda, that can improve the chances of meeting and preying. The red panda also moderately creates suitable habitat for golden snub-nosed monkey, which is conducive to the survival of golden snub-nosed monkey. (5) In the community, commensalism, mutualism, and counter anti-predation strategies of predators enhanced inter-specific associations significant. Compared with giant panda, red panda has more interspecific associations and transfers more energy.