Sally, Something which seems to be missing from the article cited is a comparison with subscription journals. Clearly, there is turnover of subscription journals just as there is turnover of open access journals. Every year, some number of new subscription journals are launched, while a significant number cease publication. I don't know whether the ratio of the number of launches to the number of closures is any different between subscription journals vs OA titles, but in principle, that could be measured. On the other hand, it still wouldn't prove a great deal. E.g. The fact that a large fraction of internet ecommerce startups failed is not an indication that ecommerce is unimportant or uneconomic. What's more important is the scale and significance of those that succeed. Ditto with OA journals. Matt