FIGURE 4 Cruise tracks for the Transient Tracers in the Ocean (TTO) North Atlantic Program, 1981.

balance a $10 million budget perfectly on a tiny calculator bought with "global change" in an airport in Tunisia.

My first impression was one of a strange trading market, where almost nothing was done by one program alone, but I saw that I had nonetheless inherited a superbly balanced program. Next I realized how fuzzy our scientist's picture was of how decisions were made. And I learned that nothing counted for more than a clear scientific question to address an unsolved problem; this made a confusing job easy. There were big programs I had never heard of and small programs of all kinds. The folding of the former IDOE big programs into the Division of Ocean Sciences had just occurred, and I began to realize that creating a new big program in this environment would be very difficult indeed. It was the new funds and separateness of IDOE that had allowed the big programs that had shaped our scientific lives to succeed. They also provided the organizational framework to make full use of our ships and yielded dense, well-populated data sets that constrained the ocean in ways not possible by other means.

I attended hearings and learned more about the need to address scientific problems fundamentally important for society. I signed off with pleasure for the publication of the long-awaited GEOSECS atlases. And I served, with Bill Nierenberg and Roger Revelle, on a very special NAS study of "Changing Climate." For 30 years Roger had kept his focus on the CO₂-climate problem. He had educated A1 Gore at Harvard, and the political world was beginning to catch up to the issue.

We had at NSF an excellent group of dedicated people: Grant Gross, Bob Wall, Curt Collins, Don Heinrichs, Bruce Malfait, Larry Clark, and rotators Mike Reeve and Rana Fine. All three rotators were to return to academia at about the same time in 1983. We met for lunch to discuss the inevitable exit interview, which we decided to do as a team. It was clear to us that the pattern we had observed, of frequent emergency requests from on high on a Tuesday afternoon for a new long-range plan by Thursday moming ("latest"), barely tolerated by the veterans, was unsustainable. Bob Wall listened, and soon after, a much more vigorous, NSF-initiated planning process took shape in several forms. In some ways it was a natural response to the vacuum created by the demise of a separate IDOE program.

The program era that followed was to prove to be fundamentally different. NSF was about to lead, with considerable courage, the new "global change" programs larger in scale and complexity, longer in planning, international in scope,