Atmospheric aerosols impose direct and indirect effects on the climate system, for example, by absorption of radiation in relation to cloud droplets size, on chemical and organic composition and cloud dynamics. The first step in the formation of Organic primary aerosols, i.e. the transfer of dissolved organic matter from the marine surface into the atmosphere, was studied. We present a molecular level description of this phenomenon using the high resolution analytical tools of Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) and nuclear magnetic resonance spectroscopy (NMR). Our experiments confirm the chemoselective transfer of natural organic molecules, especially of aliphatic compounds from the surface water into the atmosphere via bubble bursting processes. Transfer from marine surface water to the atmosphere involves a chemical gradient governed by the physicochemical properties of the involved molecules when comparing elemental compositions and differentiating CHO, CHNO, CHOS and CHNOS bearing compounds. Typical chemical fingerprints of compounds enriched in the aerosol phase were CHO and CHOS molecular series, smaller molecules of higher aliphaticity and lower oxygen content, and typical surfactants. A non-targeted metabolomics analysis demonstrated that many of these molecules corresponded to homologous series of oxo-, hydroxy-, methoxy-, branched fatty acids and mono-, di- and tricarboxylic acids as well as monoterpenes and sugars. These surface active biomolecules were preferentially transferred from surface water into the atmosphere via bubble bursting processes to form a significant fraction of primary organic aerosols. This way of sea spray production leaves a selective biological signature of the surface water in the corresponding aerosol that may be transported into higher altitudes up to the lower atmosphere, thus contributing to the formation of secondary organic aerosol on a global scale or transported laterally with possible deposition in the context of global biogeocycling.