The search engine giant, whose mantra is "to organise the world's information and make it universally accessible and useful", had been the subject of a number of lawsuits brought by publishers, authors and industry bodies, over its intentions to make millions of books searchable online.

The agreement, which still needs approval by the US District Court for the Southern District of New York, will mean better access to preview and search out-of-print (but in-copyright) books online, along with services to buy or subscribe for full access.

The newly formed not-for-profit Book Rights Registry will oversee distributing proceeds to publishers and authors.

Full access to these books from within US libraries and universities will be free.

"This historic settlement is a win for everyone," said Richard Sarnoff, chairman of the Association of American Publishers.

"From our perspective, the agreement creates an innovative framework for the use of copyrighted material in a rapidly digitising world, serves readers by enabling broader access to a huge trove of hard-to-find books, and benefits the publishing community by establishing an attractive commercial model that offers both control and choice to the rights holder."

Sergey Brin, co-founder and president of technology at Google, said: "While this agreement is a real win-win for all of us, the real victors are all the readers. The tremendous wealth of knowledge that lies within the books of the world will now be at their fingertips."

The one cloud in Google's azure sky is Harvard – although it has allowed Google to scan its out-of-copyright works, it has not signed up to the agreement, citing concerns of potential limitations on access to the books by those in higher education and those using public libraries.

Even lacking Harvard's portfolio, Google's scale and ability to match online search with information from digitised books could put a fire under the on-demand industry.