chaired by Jim Baker and carefully observed by Gerold Siedler (Kiel) as president of SCOR. Two key pieces of information came to light at this critical time. We had received the challenge to produce a satellite global chlorophyll image from compositing the CZCS fragments. Gene Feldman rose to the occasion and produced a beautiful image; we saw it for the first time in the luggage area at De Gaul airport in a gray dawn light. It made an enormous impression.

We had earlier received a scientific challenge for the proposed global CO₂ survey. In essence it was, "Show us that you can treat ocean CO₂ data in the same rigorous manner that we treat the transport of heat." This was very reasonable. By placing ocean CO₂ and heat transport in the same observational and theoretical framework, we could link the climate and greenhouse gas signals much more directly. But, as we have seen, the history of such measurement was fraught with difficulty. David Dyrssen (Goteborg) and I had drafted a position paper for the meeting on this very topic and had computed the CO₂ and nutrient fluxes across 25°N in the Atlantic Ocean. The problem was far more tractable than we had believed, and new concepts of constraining the mass balance by incorporating some adventurous organic carbon measurements had to be called upon.

It was Jim Baker who suggested that the program henceforth be called JGOFS (not Japanese, but Joint, he quipped). It took. Bernt Zeitschel became the first chair and the first JGOFS International Expedition took place in the North Atlantic in 1988.

This was my first experience at attempting a truly international effort, and I think we all found that it wasn't easy. Neil Andersen's courage, tenacity, and international experience were to serve us well throughout this period. Special mention must also go to Elizabeth Tidmarsh (now Elizabeth Gross) as executive secretary of SCOR for superb efforts in implementing the international form of the program.

Understanding the role of the ocean in the global CO₂ equation is not easy, even for many chemists. And here we had a diverse international collection of scientists of several different disciplines, many of whom were now being asked by their government agencies for informed comment on this topic as interest in greenhouse gas policy grew. We soon found that huge differences of opinion occurred. The ocean uptake of fossil fuel CO₂ from the atmosphere is not controlled by biological activity, but is an inorganic phenomenon. But the background ocean CO₂ level, which the rising trend is imprinted on, is. We are not writing the industrial signal on a blank ocean page. This detail was lost on many, and several highly contentious meetings took place. Years later, we find that a large international population of ocean scientists is now fluent in these issues, and this is a very good thing. The JGOFS program is still in place today, and the results are superb.

The transition from observation and diagnosis of the carbon cycle to active intervention by changing industrial policy, and ocean CO ₂ manipulation by disposal and/or fertilization, is about to occur and I have no doubt that NSF will provide the leadership for the scientific underpinnings needed.

What can we learn from the big program versus small program theme of this session? Firstly, the style of so-called big programs has changed enormously over the years—from the miscellany of the Indian Ocean expedition, to the large dedicated staff and many-year theme of GEOSECS, to the ensemble cast of TTO, to the remarkable coalescence of individual efforts within JGOFS to attack a very broad problem in a structured way. At each step NSF has shown leadership and creativity in crafting these efforts. And it has enabled the discovery of the fundamental pattern and time scale of ocean circulation, the invading chemical signal of the twentieth century, the chemistry of strange seas, and the fundamental basis for biogeochemical balance as we approach a warmer world. Big programs are not impersonal, but are unusually intense experiences for dozens of small groups. They are a critical part of our ocean science community.

While I have concentrated here on personal experiences, I suspect that others, in parallel programs, have similar tales to tell. The balance of small and large programs comes naturally; theory, instruments, methods, all typically come from small efforts, and the big programs cannot do without this. The best large programs embrace theory, create "small" initiatives, and provide superb opportunities for a very large number of scientists. The future is a bit more worrisome. With the desire to detect global change there is a call for very large-scale operational programs, with data continuity and massive modeling as the goal, rather than a set of evolving questions. Fortunately, there is a new class of medium-scale projects emerging, there are satellites for global observations, and there are exciting new possibilities of sensing and manipulating ocean chemistry in entirely novel ways. NSF is not a mission agency, and I hope that over the next 50 years of ocean discovery the Foundation will keep the healthy large-small ocean program balance in place.

This paper is dedicated to all my friends at NSF, over many years. It is supported by a grant to MBARI from the David and Lucille Packard Foundation.

Bradshaw, A.L., P.G. Brewer, D.K. Shafer, and R.T Williams. 1981. Measurements of total carbon dioxide and alkalinity by potentiometric titration in the GEOSECS program. Earth Planet. Sci. Lett. 55: 99-115.

Brewer, P.G., and A. Bradshaw. 1975. The effect of the non-ideal composition of sea water on salinity and density. J. Mar. Res. 33: 157-175.