NEW YORK (Bloomberg) - Google Inc. will pay $125 million to settle two copyright lawsuits by publishers and authors over its book-scanning project, a "historic" deal that the company said will make millions of books searchable and printable online.

The owner of the most popular Internet search engine said the agreement will expand the Google Book programme to let online readers search for and buy copyrighted and out-of-print books in whole or page-by-page, and provide US libraries with free access to the database.

"The tremendous wealth of knowledge that lies within the books of the world will now be at their fingertips," Google co- founder Sergey Brin said yesterday in a statement, calling the accord "historic".

The Author's Guild and members of the Association of American Publishers sued Google in 2005, claiming the book program's digitising process infringed copyrights on a massive scale. The project, which began in 2004, includes Harvard University, the New York Public Library and about 10,000 publishers in an effort to make books searchable online.

Yesterday's deal, which must be approved by a judge in Manhattan federal court, ends the lawsuits and expands what users find online when they search for a book. Searches in the Google Book program currently generate about three or four lines of text from a work. The settlement will expand the results to several pages and let readers buy full access to the content, the parties said.

Google and its former publishing adversaries will use $34.5 million of the settlement fund to create a registry program to compensate rights holders, according to court papers. Another $45 million will be used to compensate authors whose works have already been scanned without permission, the parties said.

Authors and publishers will have final say on whether their copyrighted works may be used by the programme, and thus essentially allow online purchases to compete with book sales, David Drummond, Google's chief legal officer, said.

"We've built a lot of controls for authors and publishers," Mr. Drummond said yesterday in a conference call. "If an author does not want his or her book to be in the program, they absolutely can opt out."

Under the deal, Google will keep 37 percent of revenue from online book sales and for advertisements that run next to previews of book pages, passing the remainder to the Books Registry, which will keep an administrative fee and leave the rest for the copyright holders to collect.

Google won't share advertising that runs along search results that include links to book preview pages.

The division of revenue from a subscription program for universities will be similar, the parties said on the conference call. The companies have not determined exactly how to distribute revenue under the subscription program, the companies said.

Google said the programme will create a new market for out-of- print books.

The company's measurement of readership may include the amount of time spent viewing a book, along with the genre and publication date, according to Mr. Drummond.

"It's always been our objective to have more than web content in the index, especially books," Mr. Drummond said. "Books are authoritative content. It's higher-quality content, in general, than what's on the web and it's just a fantastic thing to have for users."

In May, Microsoft Corp., the world's biggest software maker, ended a programme that let web users search through digital versions of books, ceding the market to Mountain View, California-based Google, which already has about seven million books scanned.

Google, which gets almost all of its revenue from advertisements that run alongside search results, needs to seek ways to branch out into other types of sales, said Mark May, an analyst at Needham & Co. in New York.

"It's a direction that Google will have to move towards, pursuing non-advertising ways of monetising the content they have," said Mr. May, who advises buying Google shares.