The first journals appeared in 1665: Le Journal des Scavans in Paris and Philosophical Transactions in London. They were the first publications with quality control, introducing concepts like approbation and imprimatur. Today we can see approximately 70 000 regular primary publications. Catalogues from subscription agencies like Swets, EBSCO and Blackwell list over 300 000 serial titles. Little has been done to improve accessibility and retrievability; book catalogues, whether printed on paper or electronically as OPACs (Online Public Access Catalogues), are still based on rather simple cataloguing rules. Journal literature is abstracted and indexed by so‐called secondary or A&I services: one of the most practical tools to access new articles in primary journals is Current Contents. There is no longer a library that can acquire all relevant publications, not even the Library of Congress (USA), the British Library (UK) or the National Diet Library (Japan). The number of publications seems to double every 15 years, and it would be interesting to correlate the number of active members of learned societies and the number of publications offered and accepted over the years. The only effective answer to this increasing problem seems to be the introduction of electronic publishing and communication over networks.