Peer review is not only a quality screening mechanism for scholarly journals. It also connects authors and referees either directly or indirectly. This means that their positions in the network structure of the community could influence the process, while peer review could in turn influence subsequent networking and collaboration. This paper aims to map these complex network implications by looking at 2232 author/referee couples in an interdisciplinary journal that uses double blind peer review. By reconstructing temporal co-authorship networks, we found that referees tended to recommend more positively submissions by authors who were within three steps in their collaboration network. We also found that co-authorship network positions changed after peer review, with the distances between network neighbours decreasing more rapidly than could have been expected had the changes been random. This suggests that peer review could not only reflect but also create and accelerate scientific collaboration. (C) 2019 The Authors. Published by Elsevier Ltd.