We have previously developed a DNA barcoding system, that allows us to perform lineage tracking of half a million yeast lineages as they undergo adaptive evolution (Levy et al., 2015). Using this system, we were able to determine that on the order of 20,000 lineages carrying beneficial mutations are able to establish within the first 200 generations of an experimental evolution performed by serial transfer in limiting glucose. We have whole genome sequenced several hundred of these beneficial lineages, and find that the Ras/PKA and Tor pathways are the major pathways for adaptation, and that autodiploidization confers a significant fitness effect (see Y. Li abstract). In this work, we have performed a new set of experimental evolutions, but this time in the presence of limiting nitrogen. We find that the rate of adaptation, as measured by population mean fitness, is much slower than in limiting glucose, and that the DFE consequently is made up of mutations of smaller effect, suggesting that laboratory yeast maybe better adapted to this condition than to limiting glucose. Furthermore, we find that the DFE is to some extent predictive of the population dynamics. We have also sequenced several hundred beneficial lineages, and find that the major targets of adaption in this growth condition are largely non-overlapping, despite both conditions involving serial batch growth with glucose as the carbon source.