Abstract
In designing a six-character alpha-numeric code, a combination of operational and scientific factors required two characters to be figures and four to be letters. It was predicted that when the figures were located in the fourth and fifth positions, immediate recall of the codes would be better than, when the figures were in any other positions.
An experiment was carried out which tested recall with the pair of figures in each of the five possible locations. The prediction was found to have been correct, and the underlying reasons for making it to have been substantially supported.