Journal indexing systems (JIS) are bibliographic databases that are used to search for scientific literature and for bibliometric analyses. This thesis addresses the emergence and growth of regional JIS, focusing on the Scientific Library Online (Scielo) and the Red de Revistas Científicas de América Latina, el Caribe, España, y Portugal (RedALyC) in a challenging environment in which the Web of Science (WoS) and Scopus prevail. WoS and Scopus are referred to as mainstream JIS and Scielo and RedALyC as alternative JIS. The research questions are: (1) Why did alternative JIS emerge in light of the dominance of WoS? (2) Why do researchers publish in journals indexed by alternative JIS? The research draws on the concepts of cognitive authority from information science, and universalism and particularism from the sociology of science. A cognitive authority is an information source that is credible. JIS are becoming cognitive authorities in the science communication system. Their credibility relies on their application of objective criteria to select journals (universalism). However, journal selection can be influenced by subjective criteria (particularism). The tensions between universalism and particularism suggest two scenarios for the emergence and growth of alternative JIS. A universalistic view suggests that they emerge to cover journals with low scientific impact and editorial standards. A particularistic view poses that they emerge to cover disciplinary, linguistic, and regional gaps created by biases in mainstream JIS, particularly in the coverage of WoS. The research questions were addressed through mixed methods to produce quantitative and qualitative evidence. The evidence was obtained from (1) documentary and literature reviews; (2) descriptive and correlational statistics; and (3) a case study that involved interviews with researchers in private and public universities in Colombia in agricultural sciences, business and management, and chemistry. The findings indicate that disciplinary, linguistic, and geographical biases in the coverage of mainstream JIS motivated the development of Scielo and RedALyC. The reasons for their growth have been conceptualised in this thesis as: (1) training; (2) knowledge-gap filling; and (3) knowledge bridging. This thesis addresses a significant gap in the sociology of science by studying new authorities in the science communication system. It contributes to debates on universalism and particularism, showing that both are involved in the selection of journals by JIS. It also contributes to understanding how particularism in mainstream JIS can pose barriers to the communication of scientific knowledge that has the potential to address pressing social demands. The findings could contribute to the design of research policy and research evaluation in contexts not widely covered by mainstream JIS.