The present investigation is based on the separation of one of the most complex food matrices: the roasted coffee bean volatile fraction. Analysis of the two main species of coffee (Arabica/Robusta) was achieved through an effective and simple sampling procedure, headspace solid-phase microextraction (SPME), and the unprecedented resolving power of comprehensive gas chromatography (GC x GC). The combination of these two techniques proved to be a powerful tool for the extraction and separation of coffee volatiles. In fact, thousands of compounds that play various roles in the constitution of coffee aroma profile were resolved in the 2-D contour plot, each occupying a specific position pinpointed by two retention time coordinates. The potential use of this method for the assessment of coffee quality and the detection of commercial fraud is discussed. The potential of GC x GC for identification and classification of unknowns was also demonstrated, as the formation of characteristic patterns for structurally related compounds was observed in the bidimensional chromatogram. Moreover, reproducibility results were supported by the use of an autosampler for SPME applications that allowed any inaccuracy arising from manual handling to be avoided.