A few days before Secretary of State Colin L. Powell gave his 2003 presentation to the U.N. Security Council on Iraq's purported weapons of mass destruction -- with its startling allegation that four individuals had confirmed that Iraq had mobile biological weapons laboratories -- a government analyst who had read a draft of the speech sent an urgent e-mail to his boss.

All those sources are suspect or unreliable, especially the key one nicknamed "Curve Ball," warned the analyst, the only U.S. intelligence official who had met Curve Ball.

The analyst received a dismissive reply. "This war's going to happen regardless of what Curve Ball said or didn't say, and . . . the Powers That Be probably aren't terribly interested in whether Curve Ball knows what he's talking about," replied the deputy chief of the CIA's Iraq task force. The warning was never passed on to Powell or his top aides.

The incident, detailed in the Senate Select Intelligence Committee's report on prewar intelligence, underscores a central theme in the committee's investigation: Although there is little evidence that intelligence analysts were pressured to change their findings, the agency ignored or belittled inconvenient or contradictory facts in its rush to present the most dramatic case against the Iraqi government.

This left the administration relying on shaky or dubious accounts from Iraqi defectors, such as the four sources who provided what Powell has called the "most dramatic" part of his 90-minute U.N. speech.

Although Powell had told the United Nations that "every statement I make today is backed by sources, solid sources," the report concluded that much of the information the CIA provided for Powell's speech was "overstated, misleading or incorrect."

Whether unusual pressure on intelligence analysts resulted in faulty reports has been one of the big questions hanging over the failure to find weapons of mass destruction in Iraq. Officially, the committee will examine how the administration used the intelligence it received in a second report, to be completed after the presidential election.

The 440-page report released yesterday concluded that key officials, including Vice President Cheney, did not unduly pressure analysts to change their conclusions. The committee said it could find no evidence that administration officials "attempted to coerce, influence or pressure analysts to change their judgment," and added there was no evidence that Cheney's repeated visits to the CIA were attempts to pressure analysts or even were perceived as such.

But the report provides evidence of how administration officials held preconceived notions that appear to have affected how they viewed the intelligence information they received.

One senior intelligence officer in the Defense Intelligence Agency told the committee that some officials working for Defense Secretary Donald H. Rumsfeld were convinced that Iraqi President Saddam Hussein was behind the 1993 World Trade Center attack and may even have been involved in the Sept. 11, 2001, attacks.

"They did not tell us what to say," the intelligence officer told the committee. "There wasn't pressure in that sense. But you certainly had to make sure that your analysis was on target and that you were very precise in the words that you used."

At one point, the committee said, Undersecretary of Defense Douglas J. Feith established a team that reviewed intelligence on the links between Iraq and al Qaeda. Feith told the committee that the team "found some things that looked very interesting in the year 2002 that apparently didn't register with people or were not given great prominence." Feith's team put together a briefing for Rumsfeld and, later, for CIA Director George J. Tenet, and then attended a meeting of intelligence analysts in August 2002 that sought to assess Iraq's links to terrorism.

The analysts told the committee they found the presence of policy staffers at the meeting unusual. "Most of the [policy] staffers' concerns had to do with the use of too many caveats to the reporting, and the 'tone' of the document," the report said, though those concerns did not change analytical judgments.

Anthony Cordesman, a national security specialist at the Center for Strategic and International Studies, criticized the report for failing to investigate the question of pressure on analysts more deeply. "I know of DIA senior officers pressured out of the Pentagon, and younger analysts who left the community over political pressure," he said in an e-mailed assessment. He said that the committee also failed "to examine the fact that the intelligence community almost always responds to the user's demands and perceptions."

The committee's report said that it made repeated calls for analysts to step forward and reveal whether they were pressured by administration officials, and that the committee sought out and interviewed analysts who had been identified in news reports. Bruce Hardcastle, a senior DIA official for Middle Eastern affairs, was interviewed after a report in The Washington Post said he avoiding meeting with Deputy Undersecretary of Defense William J. Luti because he sharply disagreed with Luti over the imminence of the threat posed by Iraq.

But Hardcastle denied the Post account, telling the committee his dispute with Luti was not over Iraq but concerned the use of the word "assassination" to describe the killings of terrorist leaders by Israeli Defense Forces. He said he did not experience pressure to change assessments on Iraq, but he added: "Generally it was understood how receptive [Defense policy officials] were to our assessments and what kind of assessments they would not be receptive to."

Secretary of State Colin L. Powell said last year, "Every statement I make today is backed by . . . solid sources." That data have been deemed "misleading or incorrect."