service fellowship programs. Even the Department of Education fellowships were increased last year to match NSF fellowship stipends currently at $14,000 annually.

The Minority Access to Research Careers (MARC) Program has proven to be highly effective in drawing talented minority students into careers in biomedical research. The program should be strongly supported and perhaps increased in size. According to the NIH, in 1990, over 5,000 doctorates were awarded in the life sciences in the U.S., but only 2.3 percent went to Hispanics, 1.4 percent to African Americans, and 0.2 percent to Native Americans.

Many universities have used the MARC program to identify promising undergraduates and provide them with summer research internships and related activities to interest them in and prepare them for graduate study. The quality of the research program and the extent of faculty mentoring are major determinants in the success of institutional programs, and these attributes should continue to play a determining role in awarding grants in any expanded program. In addition to expanding the MARC program, the committee should also consider opportunities to expand and increase the flexibility of incentives for principal investigators to hire minority graduate and undergraduate students.

While women now receive about one-third of the Ph.D.s granted, they still remain underrepresented in science and engineering. Mr. Chairman, I would encourage this Committee to consult with the Office of Research on Women’s Health at the NIH to explore ways in which the NRSA program might encourage more women to consider a career in biomedical research.

Finally, one of the features of the NRSA training grant that might be strengthened to assure the maintenance of high quality research environments is the need for institutional flexibility in allocating the resources that accompany each grant. In addition to providing critical support for students, training grants should continue to provide significant resources for strengthening and sustaining the training programs themselves. The Committee may wish to consider permitting a flexible rather than a fixed proportion of training grants to be used for program support. However, care must be taken to assure that any increased proportion of training grants used for program support be fully justified on the basis of the overall benefit to training capacity of the program, and clearly differentiated from objectives better supported through research funding.

To increase NRSA stipends to a competitive level while maintaining the same number of trainees in the program will require a substantial increase in program funding. I fully understand how difficult this will be, given the constraints on the NIH budget. I suggest, therefore, that you phase in the necessary increases over perhaps a five-year period.

There are three areas I want to address to assure a continuing supply of skilled investigators in the biomedical and behavioral sciences: 1) the need for interdisciplinary and team research including multi-university relationships; 2) increased attention to the educational needs of women and minorities; and 3) strengthening financial support and allowing more flexibility regarding employment stipulations.

First I will address the need for interdisciplinary and team research and cross university relationships. My recommendations here are three fold: 1) more emphasis on cosponsors and interdisciplinary sponsors; 2) financial support for consultants from smaller or lesser known universities; and 3) some sort of “match” system with faculty across disciplines and universities.

The need to mentor younger and newer scientists is clearly critical and, as funding becomes increasingly competitive, senior researchers find that they have less and less time to devote to mentoring. Senior researchers and faculty need both time and rewards for mentoring younger researchers. One alternative may be to share the mentoring when that is possible.

For example, one of the strengths of my NRSA was that I had not only two cosponsors from The University of Michigan--each from a different discipline and with complementary skills, but also I was able to utilize the mentorship of a research team at another university, Michigan State University, because of my employment there as a research associate. My mentoring occurred across disciplines and across universities. This cooperation is unusual I am told; yet, this broad perspective was mutually beneficial at every level. The responsibility for mentoring was shared, and each one of us had the benefit of interaction with other experts in my particular field of research.

Cosponsors need to be the rule rather than the exception. The traditional model of a solo researcher