Adults sometimes report highly specific details of childhood events, including the weather, what they or others were wearing, as well as information about what they or others said or were thinking at the time. When these details are reported in the course of research they shape our theories of memory development; when they are reported in a criminal trial they influence jurors' evaluation of guilt or innocence. The key question is whether these details were encoded at the time the event took place or have been added after the fact. We addressed this question prospectively by examining the memory accounts of children. In Experiment 1 we coded the reports of 5- to 6-year-olds and 9- to 10-year-olds who had experienced a unique event. We found that spontaneous mentions of these specific details were exceedingly rare. In Experiment 2 we questioned additional children about a similar event using specific questions to extract those details. We found that 9- to 10-year-olds were able to accurately answer, while 5- to 6-year-olds had considerable difficulty. Moreover, when the younger children did respond they provided generic, forensically inadequate, information. These data have important implications for the courtroom and for current theories of memory development and childhood amnesia.