The Oklahoma County (Oklahoma) Library executed a six-month experiment in 1974 on one of its five bookmobiles to determine whether: (1) patrons would accept a primarily paperback book collection; (2) the cost of unreturned paperbound books in a no-fine and no-overdue-notice situation would be cost-effective when compared to the predictable loss of hardbound books and the resulting cost; (3) a category-arranged collection (rather than a cataloged one) would prove useful and acceptable to patrons. Paperbacks were proven to be even more acceptable than hardbacks, except that patrons did want access to bestsellers available only in hardback. In spite of a higher number of books unreturned, the dollar loss per 100 circulation was lowered by the use of paperbound books. It was considered practical, however, to send overdue notices to those with a large number of unreturned books. A circulation study showed that the categorical arrangement appeared to be acceptable to patrons, but some redesign of the topical categories should be considered. The success of the experiment led to use of the same system on the other four bookmobiles. (LS)