The history of irrigation organizations during the late nineteenth and early twentieth centuries in the arid and semi-arid American West reveals the important role of experimentation in determining the institutional forms that evolved. The legal framework was such that a wide variety of bottom-up organizations developed to deal with transaction cost problems. Contracting was complicated by asset specificity, potential spillovers between users, 0-riding, and holdout problems. Asset specificity on the part of both farmers and irrigation infrastructure owners should have led to vertical integration with a single firm owning both the farmland and the infrastructure. However, the differing economies of scale between capturing and delivering irrigation water and farming meant that vertical integration would have resulted in costly operations. Instead of vertical integration or the separate ownership of farms and irrigation infrastructure, western farmers sought out intermediary institutions. In some cases, simple contracts among a few farmers were sufficient to divert water from streams and to carry it to crops. In other cases, a developer would form a commercial irrigation company, buy a large block of land, install irrigation infrastructure, and then sell off farms. However, the commercial irrigation company did not provide solutions to transaction cost issues; instead, it was fraught with transaction cost problems that undermined its usefulness. Mutuals, both incorporated and unincorporated, allowed farmers to contract with other farmers to own and operate irrigation facilities. Irrigation districts, a form of localized government with coercive powers, were authorized in all of the western states. These districts became another significant form of irrigation organization.