Autobiographical memories play a significant role in shaping a sense of self and are essential for normal functioning on a daily basis. Music-evoked autobiographical memories (MEAMs) are memories of past events that are triggered by music, often involuntarily. In this study, we had two aims: (1) to explore the qualitative characteristics of MEAMs in healthy aging and (2) to analyze possible relationships between types of MEAM and the emotions evoked by music. We played 140 excerpts from songs from the four decades between 1950 and 1990, and folk songs, to 82 healthy adults aged 60–86 years and collected their MEAMs, expressed in their own words. We applied content analysis to the data and categorized them manually. Participants rated the valence, arousal, and intensity of emotions evoked by each excerpt. We conducted a mixed-effects logistic regression analysis to identify relationships between participants’ MEAMs and ratings. Participants frequently reported music-related activities such as singing, dancing, and listening, and specific people and places, in their MEAMs. The excerpts evoked more autobiographical memories from participants’ youth than from any other periods in their lives. Our analysis of the relationships between types of MEAM and the emotions evoked by music suggest a variety of emotional profiles. The choice of methods emphasized ecological validity and the findings provide a new perspective on the nature of MEAMs in healthy aging.