THE CONNECTION between scientific research and a vigorous economy is hard to measure but impossible to question. Though the payoffs from basic research are unpredictable--Alexander Fleming could never have guessed that in the space of a few years his interest in a curious green fungus (penicillium) would transform mankind's life expectancy--it is predictable that payoffs will come. A strong research enterprise reliably produces technological advance, innovation and productivity growth.

The more than decade-long decline in U.S. investment in research and development, therefore, is --or, rather, should be--a matter of deep national concern. American science is still very strong, but since about the middle '60s investment in research in proportion to the GNP has dropped sharply. The drop was evident in both government funding and private investment. By many measures, technological growth has followed suit.

The administration's emphasis on cutting federal spending in order to rejuvenate the economy could not have come at a worse time for science and engineering. Nor could it be less likely to produce the desired result. Although the administration made some effort to shield basic research funds--as distinct from applied research and development--from its budget cuts, it did not do nearly enough. The 1983 budget proposes sharp declines in real research dollars for most agencies. Research and development in the fields of energy, agriculture, environment, transportation, commerce and health and in the disciplinary research programs funded by the National Science Foundation are all down. Those national laboratories not primarily devoted to weapons research are absorbing devastating cuts and disruption to their most productive research programs.

The only exceptions to this trend are defense and space programs, the latter largely because of their military applications. The 1983 budget proposes a 25 percent increase above inflation for defense R&D in two years. If Congress were to approve these levels, military spending would consume nearly two- thirds of all federal research funds. Though some of the Pentagon's riches will be usefully spent rebuilding the country's engineering schools and laboratories, that is still a dangerously lopsided allocation of research effort.

The private sector cannot pick up the slack from federal cuts in research. If industry were to triple its investment in university-based research and sustain that level for a decade, it would just about offset the proposed federal R&D cut for a single year. Nor is such research industry's responsibility. Enlightened self-interest should lead business to focus more on future competitiveness and less on next year's bottom line. But government has a strong role in the society's economic future, and that means more dollars and more consistency in the support of R&D.