The interval between two presentations of the same item has a marked effect on paired associate learning (e.g., Greeno, 1964) but has been reported to have no effect on free recall (Waugh, 1963). To explore this apparent contradiction, a free recall experiment was performed with a mixed list of words and nonsense syllables in which the interval between two presentations of the same item was varied. A significant spacing effect was obtained for nonsense syllables but none for words. It is suggested that spacing may be important only for new associative or response integrative learning.