The last few years there has been a growing number of body-wave observations in noise records. In 1973 Vinnik conjectured that P-waves would even be the dominant wavemode, at epicentral distances of about 40 degrees and onwards from an oceanic source. At arrays far from offshore storms, surface waves induced by nearby storms would not mask the body-wave signal and hence primarily P-waves would be recorded. We measured at such an array in Egypt and indeed found a large proportion of P-waves. At the same time, a new methodology is under development to characterize the lithosphere below an array of receivers, without active sources or local earthquakes. Instead, transmitted waves are used which are caused by distant sources. These sources may either be transient or more stationary. With this new methodology, called seismic interferometry, reflection responses are extracted from the coda of transmissions. Combining the two developments it is clear that there is a large potential for obtaining reflection responses from low-frequency noise. A potential practical advantage of using noise instead of earthquake responses would be that an array only needs to be deployed for a few days or weeks instead of months, to gather enough illumination. We used a few days of continuous noise, recorded with an array in the Abu Gharadig basin, Egypt. We split up the record in three distinct frequency bands and in many small time windows. Using array techniques and taking advantage of all three-component recordings we could unravel the dominant wavemodes arriving in each time window and in each frequency band. The recorded wavemodes, and hence the noise sources, varied significantly per frequency band, and -to a lesser extent-per time window. Primarily P-waves were detected on the vertical component for two of the three frequency bands. For these frequency bands, we only selected the time windows with a favorable illumination. By subsequently applying seismic interferometry, we retrieved P-wave reflection responses and delineated reflectors in the crust, the Moho and possibly the Lehmann discontinuity.