The role of anonymity in giving is examined in a field experiment performed in thirty Dutch churches. For a period of 29 weeks, the means by which offerings are gathered is determined by chance, prescribing for each offering the use of either 'closed' collection bags or open collection baskets. When using baskets, attendants' contributions can be identified by their direct neighbors, and attendants can observe the total amount given by the people who preceded them. Initially, contributions to the services' second offerings increase by 10% when baskets are used, whereas no effect is found for first offerings. The positive effect of using baskets peters out over the experimental period. Additional data on the coins collected show that in both offerings, people switch to giving larger coins when baskets are used.