The numbers will not surprise anyone who has closely tracked federal budget discussions about science and technology in recent years. But that won't make the data released Friday by the National Science Foundation any more palatable for those concerned about the American research enterprise.

The science foundation's Division of Science Resources Statistics, the preeminent federal source of data about research in the United States, issued its first look at information about the 2007 fiscal year from its annual survey of academic research and development. The survey examines what colleges and universities report spending on scientific and other R&D, showing actual dollars on the ground as opposed to what Congress appropriated or federal agencies planned to allocate.

The data largely confirm the reality, evident from reports by the American Association for the Advancement of Science and other entities that monitor federal spending, that federal support for science has slowed considerably from the heady days of the early part of this decade, when Congress completed an effort to double spending on biomedical research.

But the extent of the downturn is stark, the NSF reports: For the second year in a row, federal funds for academic research in science and engineering failed to outpace inflation. That situation is "unprecedented," the science foundation notes; never before in the 36 years that the NSF has produced this report have there been two consecutive years in which federal science spending has declined in current dollars.

As seen in the table below, colleges and universities spent $49.431 billion on research and development in the 2007 fiscal year, up 3.5 percent (or 0.8 percent in inflation-adjusted dollars) from 2006. Federal support rose to $30.441 billion, up 1.1 percent in current dollars and down 1.6 percent once adjusted for inflation. That follows a 0.2 percent decline in inflation adjusted dollars from 2005 to 2006:

College and University R&D Expenditures, Fiscal 2002-7

The dip that these data confirm in federal support, which is by far the biggest source of money for academic research, deeply troubles college and university leaders. "Our global competitiveness depends on America’s having the best science and technology," Joel Seligman, president of the University of Rochester, said in an e-mail. "To keep our lead, we must support it. It is vital that federal funding at least keep pace with our closest competitors in Europe and Asia."

The impact is increasingly felt on campuses, Mark Wrighton, chancellor of Washington University in St. Louis, said in a telephone interview Saturday. "We're seeing really talented scientists lose their grants, not because they don't have strong proposals, but because the grant money isn't there. To say it's demoralizing is an understatement. It deters postdocs and over time will erode our ability to have new applications from basic research."

As the data above show, universities are turning to other sources to try to make up for the disappearing federal money. State and local government backing grew by 6.1 percent, and funds from other sources -- nonprofit groups and other nongovernmental entities -- rose by 10 percent. Industry-sponsored research, which dropped from 2001 to 2004, has increased since then and grew by 11.2 percent from 2006 to 2007. But that of course has its own issues, as concerns intensify -- not least in Congress -- about potential conflicts of interest as scientists depend more and more heavily on corporations for support.

Universities also tapped into their own endowment and gift funds to try to fill the gap. "Universities have had to dip into endowments or tuition or internal funds to keep the labs' lights on if investigators are falling short in research grant competitions," said Kei Koizumi, director of the R&D Budget and Policy Program at the American Association for the Advancement of Science. "Or they are using their own funds to get the labs equipped or built so they can better compete for increasingly scarce federal funds."

"We are calling on private resources to provide more bridge funding for investigators," said Wrighton of Washington University, which reported spending $573 million over all on academic R&D in 2007, up from $530 million, which placed it 18th on the NSF list.

Despite widespread agreement -- pleaded for in a stream of recent reports and prescribed in federal legislation -- on the need for the country to bolster its spending on the physical sciences, the NSF data show that financial support for the physical sciences is inconsistent. The table below, which shows how individual disciplines fared in federal support in 2007, reveal that funds for the physical sciences fell vis-a-vis inflation, the life sciences rose slightly, and engineering fields, comparatively, thrived. Certain individual fields, such as oceanography and mathematical sciences, also did pretty well:

Wrighton and other university leaders acknowledge that the odds of altering the current trend line for federal support for academic research may be long, given the intense competition for increasingly scarce resources -- especially in the short term, as the American economy seeps into what may be a recession. But they are hopeful that November's election, whoever is elected, offers at least the promise of a new context for the conversation.

"Unfortunately we have not made the case [for increasing federal backing of academic R&D] strongly enough so far," said Wrighton. "But we have a tremendous opportunity to work with the campaigns. Each of the campaigns has been emphasizing the need to strengthen the American economy, and we obviously believe that doing that is tightly linked to advancing the sciences."

Below are tables that show the institutions that spent the most on academic R&D over all and in several subcategories in the 2007 fiscal year. Over all, the top 20 institutions accounted for about 30 percent of all university spending. The biggest change from 2006 to 2007 was at Duke University, which the NSF attributed to a hefty increase in funding from industry as well as from the National Institutes of Health.

Top 20 Institutions in Academic R&D Spending, 2006-7

Academic R&D, Top 20 Institutions Without a Medical School, 2007

R&D Expenditures, Top 20 Minority-Serving Institutions, 2007