The effectiveness of two memory training programs designed to enhance four-digit number recall was examined in 90 healthy older adults. One group received instruction and training in the number-consonant mnemonic, whereas another group was instructed to adopt their own encoding and retrieval strategies to enhance number recall. Also, a control group receiving no training between testing occasions was included. The criterion task was administered according to the Buschke selective reminding procedure. Posttest performance was evaluated with and without cognitive support for remembering (i.e., verbal cues). Under unsupported conditions, the mnemonic group improved number recall following training and the selfgenerated strategy group showed a tendency in the same direction. When support was provided, group differences in favor of the two training groups increased. In addition, no training-related gains were observed in two verbal transfer tasks. The relatively similar patterns of gains in the two intervention groups were discussed in terms of advantages and disadvantages in the two training regimens balancing each other.