In this project two major field demonstrations were carried out in conformity with the goals of the Class Revisit Program. The first demonstration was at Vernon Field in Isabella County Michigan, and the second was the Springdale demonstration in Manistee County, MI (Figure 1). The Vernon Field project targeted the Devonian Dundee Formation, a prolific producer in Michigan while the Springdale project targeted the ?Brown Niagaran? an informal stratigraphic unit in the Lower Silurian (Figure 2). The first well at Vernon Field, the State Vernon & Smock #13-23 (Permit: 53945), was spudded on October 12, 2000 but was not economically successful, even though an extensive geochemical program was carried out. A second well, the Bowers 4-25 HD (Permit: 54950), was drilled to test the geochemical anomaly, but was also plugged and abandoned on December 3, 2002. No further wells were drilled at Vernon Field. The Springdale demonstration was much more successful, current production exceeds 257,000 BBLS/oil and 1,600,000 MCF gas, and demonstrated what appears to be the pattern for the future in Michigan and by extension to other mature gas and oil provinces. Specifically, Springdale is going to be a million barrel field, counting both oil and gas equivalent, and is probably one of many such fields in the Michigan Basin that will be discovered and exploited in the future. The key technologies employed at Springdale were 2D seismic and horizontal drilling. The horizontal drilling essentially probed the formation for porosity and hydrocarbons, was then used for production, and is scheduled for follow-up flooding duty. This type of play is typical of many of the recent plays in Michigan in that they are conceived, planned and conducted by a relatively small company and consultants. A rough rule of thumb is that probably no more that 6-12 professionals, including landmen and management, were/are involved in these plays and the key is very low overhead, with much of the risk spread over 5-10 investors. The DOE investment (approximately $2 million) has been timely and significant; both Vernon and Springdale relied on DOE funding in the early, critical stages. We estimate a total (gross) yield of $40 ? 60 million, based on a total Springdale production of 1 million barrels of oil and oil equivalents, with over $5 million paid back to State in royalties and lease fees. Vernon Field was instructive too, since it demonstrated the utility of surface geochemistry and provides a type case for exploitation of mature fields. The initial geochemical surveys were consistent with the later, very comprehensive survey (>1200 samples) and consistently indicated that Vernon was not a prime target. In retrospect, we should have conducted a comprehensive survey at the beginning, or at least established a regional baseline in order to better access our results. The lesson learned is two-fold: don't skimp on the geochemical surveys upfront and walk away if they are not optimistic. Perhaps, that should be very optimistic. The difficulty is that an industry used to making decisions based primarily on seismic data and interpretation is reluctant to invest in another technology, particularly one that is unproven in their eyes. Nevertheless we see a bright future for surface geochemistry in the exploitation of mature basins provided that two things develop: one is better technology, perhaps based on ?sniffer? approaches which permit real-time data collection and assessment, and, two, lower costs per sample. These are not unrelated; field-based data collection and interpretation will eliminate lab costs, which can be 50-70% of total survey costs. In this respect, the geochemical industry seems to be working at cross purposes: lab costs increased on the order of 50-100% over the 5-year time span of this project.