If anything is growing super-exponentially (where the rate of growth also grows) it is the annual number of scientific publications. Since the number of potential readers is not, a crunch point is surely coming where the average number of readers of a learned paper may be just the authors themselves: someone with the time can do the necessary arithmetic to check that casual prognosis. To a semi-professional browser of a few leading journals in one subject area, it does sometimes seem as if that point arrived a while back. Yet the number of actual journals is growing as well, for example Nature now has more than 30 satellite journals, when once it stood at stratospheric height astride the entire breadth of Science. One journal, which professional etiquette will not allow me to mention by name, is currently submerged beneath a backlog of unpublished but accepted papers, to the extent that a decision to jump from 18 issues to 24 per year has run into difficulties. Its next 6 issues will be bundled into two weighty volumes to clear the desks of a single sub-editor, who it seems is definitely getting a bit ‘frayed around the edges’. How do publishers manage to blurt out such an awesome volume without a sort of heat death of the literary universe and, more mundanely, economic collapse? For a start, they take adverts and up the subscription rates, both for paper and on-line versions. Oddly there is a huge range of subscription rates for top-ranking geoscientific journals, from about £100 for the beautifully typeset and edited Journal of Geology published by University of Chicago Press to almost £3000 for Palaeogeography, Palaeoclimatology, Palaeoecology published by Elsevier (2005 rates). Publishers can charge for colour content – most do but a few do not. Of course, they do not pay for a paper’s creative content, unlike almost every other kind of publication, but that is born of hubris, which is probably inbred among scientists. For the same reason they do not pay their referees; in fact I am sure that some would pay for the privilege (see above). Let us be clear on one thing: scientific publishing is a hugely profitable business, and most scientists would be hamstrung if they did not go along with its rapacity, unless perhaps they club together to create free on-line, peer-reviewed journals outside the pack.

Among the ruses aimed at milking the demonically possessed Gadarene herd, to which most scientists belong, is one worth special approbation: compulsory assignment of copyright (Marris, E. 2006. PS I want all the rights. Nature, v.442, p. 118-119). That is relatively new, and along with it is a rapid attrition of the only perk, free offprints to distribute, amaze your friends and rub salt in the wounds of your enemies. It is now common to give authors only the final on-line PDF, file, but with stern warnings that only a certain number of free distributions are allowed. To get paper offprints (still much more highly valued by colleagues than electronic text), authors increasingly have to pay extortionate fees, almost equivalent page per page to the market value of a hand-illuminated, mediaeval manuscript. Let’s get this straight, publishers do not pay authors (and referees), and often demand payment for necessary colour work that visually enhances journals (costs of 4-colour printing have plummeted in recent years), yet claim ownership of the published article, effectively violating the intellectual property rights of authors. A powerful move against this is beginning. Original PDFs of almost 50 % of papers inScience and Nature and increasing numbers of those in other prestigious journals are appearing for free download from the Web. Major grant givers in the medical sciences now demand that work that they sponsor becomes free to all, once published. Some provide authors with forms to add to the copyright transfer papers that must be signed to ensure free access to all. Publishers complain direly that such action will destroy their journals financially, claiming that, ‘the final version is where publishers add value’ and that the version published by a journal is ‘definitive’, ‘part of the minutes of science’. Readers can draw their own conclusions about this issue of form versus content. It is relatively simple to prepare PDFs of corrected papers and their figures to a publishing standard and distribute them freely, a widely adopted move among physicists, biologists and medical scientists through open-access libraries such as PubMedCentral.