Harvard researchers said yesterday they have asked the university's ethics boards for permission to create cloned human embryos for medical research, marking the first push to conduct such experiments at a U.S. academic institution since a failed attempt in 2001.

The goal of the ethically contentious, privately financed work -- which has already gained provisional approval from one Harvard committee -- is to develop new cures for diabetes, Parkinson's disease and other ailments. The approach involves creating cloned human embryos that would be destroyed within days in order to retrieve stem cells growing within.

Several lines of evidence suggest stem cells from cloned embryos have greater potential as medical treatments than stem cells derived from unused embryos at fertility clinics, which are created by in vitro fertilization and are now the major source of stem cells for research.

Opponents, however, say it is wrong to create human embryos solely for the purpose of destroying them. Some also fear that the work could speed the arrival of the first cloned baby -- an outcome that virtually all parties to the debate expressly oppose.

The only previous reported effort to produce stem cells from human cloned embryos at an American university was approved at the University of California at San Francisco but ended in failure. The Harvard move comes at a politically precarious time, as stem cell research has emerged as a potent wedge issue among voters in the razor-close presidential election.

President Bush has encouraged Congress to pass legislation banning all human embryo cloning, and the House has repeatedly voted to do so. The Senate has remained split on the issue.

Democratic challenger Sen. John F. Kerry has said he supports research on cloned embryos with proper federal and institutional oversight.

The United Nations, too, is embroiled in the issue. On Oct. 21, it will renew a twice-stymied effort to gain consensus on whether there should be an international ban on creating cloned embryos and cloned babies, or only babies. Some countries have already said they would not abide by a broader ban.

And next month, Californians will vote on a ballot initiative that would provide millions of dollars for stem cell research, potentially including work involving human embryo clones.

The Harvard application highlights an evolving migration of U.S. human embryonic stem cell research from the federal research arena into the far less-regulated private sector as scientists grow frustrated by restrictions on government funding.

An executive order signed by Bush in 2001 allowed limited use of federal funds for human embryonic stem cell research and banned federal support for research involving human embryo clones.

Privately funded research facilities, including a few that have recently worked out partnerships with universities, are allowing American scientists to compete aggressively with their counterparts in the growing number of countries that allow such experiments.

South Korean scientists earlier this year became the first to derive stem cells from a cloned human embryo (a Massachusetts company previously reported it had made cloned human embryos but was unable to get stem cells from them). England recently gave researchers there permission to do the same, and other countries are considering such approvals.

Coincidentally, the National Academy of Sciences yesterday completed a two-day meeting in Washington aimed at generating voluntary guidelines for private entities conducting such research in this country. While participants in that meeting were mostly supportive of cloning research, some spoke out vociferously against it -- among them Leon R. Kass, who chairs the President's Council on Bioethics.

Therapeutic cloning, as the practice is often called, "crosses a new and major moral line," in part because it represents a "purely instrumental use of human embryos," said Kass, who said he was speaking for himself and not the council.

Cloned embryos are created by fusing a single skin cell from the person or animal to be cloned with an egg cell that has had its own genetic material removed. Under proper conditions, that combination can be made to grow into an embryo that is genetically identical to the original cell donor.

The Harvard proposal, submitted by stem cell researcher Douglas A. Melton, calls for the creation of human embryos that would be genetic replicas of various patients with Type 1 diabetes, Parkinson's disease and Alzheimer's disease, said Charles Jennings, executive director of the Harvard Stem Cell Institute, where Melton's work would be done. The institute was formed earlier this year.

Stem cells retrieved from those embryos would carry the exact collection of genetic glitches -- many of them as yet unidentified by scientists -- that cause those diseases. By watching what goes wrong with those cells as they mature in laboratory dishes -- into brain cells, for example, in the case of Parkinson's -- scientists hope to discover underlying causes of the diseases and test drugs to see whether they can correct the problems.

A second Harvard effort, not yet submitted for review, calls for the creation of embryos that are clones of patients with various blood disorders, such as sickle cell anemia. One goal of that research, to be led by Leonard I. Zon and George Q. Daley of Children's Hospital Boston, a Harvard affiliate, is to correct the disease-causing defects in stem cells harvested from those embryos, with an eye toward someday gaining government permission to infuse the corrected cells back into patients.

Experiments in mice indicate that the cells would settle into the patients' bone marrow and start producing a steady supply of healthy blood cells, curing the patients. And because the cells would be derived from embryos that were clones of the patients themselves, they would be genetically identical to each patient and not rejected by the immune system.

Jennings said Melton's plan was reviewed first by a special committee of scientists, ethicists and others set up to deal expressly with stem cell proposals. That committee, which answers to Harvard's provost, recently approved the proposal in principle, he said.

The plan must still gain approval from several "institutional review boards" at Harvard, at the fertility clinic at which women would be asked to donate the needed eggs, and at the clinic or hospital responsible for the patients who would donate their cells to create the clones.

"Harvard has certainly recognized that this raises a number of issues, certainly a number of them controversial," Jennings said. "We hope that many points of view will be aired in the discussions to come."

Daley said his team, which would also work at the stem cell institute, has had "extensive meetings" with officials at Children's Hospital, including researchers, ethicists, lawyers, financial and intellectual property specialists, "to make sure that everyone's concerns are addressed."

He would not predict when his proposal would be submitted for review.