WWW- Sources close to the defense in the Duke University rape investigation deride the case as “pin-the-tail-on-the-donkey justice” – and now a police report reveals how, by that analogy, the tail was pinned on two Duke lacrosse players now accused of rape.

A 15-page document shown to Darla Miles of WTVD, an ABC-owned station in Durham, N.C., described how the alleged rape victim, a 27-year-old exotic dancer and mother of two, identified three lacrosse players as those who she said attacked her the night of March 13.

According to the police report, the alleged victim was shown a police lineup of 46 photos individually depicting all the Duke lacrosse team members except for freshman goalie Devon Sherwood, the only black member of the team. He was excluded because the alleged victim told police her attackers were white.

After being shown the pictures in a sequence of PowerPoint slides, the document adds, the woman said she could identify the two players indicted April 17 with 100 percent certainty. She picked out Reade Seligmann as the attacker who forced her to perform oral sex and Collin Finnerty as the second man to rape and sodomize her.

She said she also could identify with 90 percent certainty the first man who raped and sodomized her. This attacker has not been arrested as of today, though District Attorney Mike Nifong said at the beginning of the week that he was looking to make a third arrest.

Nifong declined ABC News requests to comment on the information in the police report – though an outside prosecutor not involved in the Duke case suggested the accuser’s photo identifications may be effective.

However, an eyewitness identification expert believes the police lineup procedure was flawed because no non-lacrosse players were included.

Gary Wells, president of the American Psychology-Law Society, described it as “a multiple-choice test without any wrong answers.”

By including “fillers,” or non-suspects, in a police lineup, an accuser has to pick past the filler to choose people who actually might have committed the crime.

“Without fillers as a control, the process has no internal credibility check,” Wells said.

David Rudolf, a North Carolina defense lawyer who has been an adjunct professor at Duke and the University of North Carolina-Chapel Hill, believes the procedures may be problematic to the point of being inadmissible in court.

“I have significant doubt that this will be admitted in court,” he said, “and no doubt defense will challenge it vigorously.”

The issue, Rudolf explains, is that due process prohibits evidence from lineups that are unnecessarily suggestive and conducive to mistaken identity.

“When you take the only suspect group and put it in front of the victim,” Rudolf says, “by definition you’re suggesting it was one of the 46 people in that group.”

How could the lineup have been done differently? Police could have used pictures of male Duke students to construct 46 lineups – each one with a lacrosse player and five six students with similar features who are not lacrosse players. From those lineups, one could determine whether the victim picks out those who allegedly attacked her.

“At that point, you have a lineup that’s not unnecessarily suggestive,” Rudolf says. “You’ll have students who are clearly not at the party.”

Defense attorneys had their own criticism about how the lineup was done and have said they would suppress the results of the lineup. Sources close to Seligmann also point out that he and other members of the team look alike, casting doubt on the alleged victim’s ability to single out her alleged attacker.

Despite questions and criticisms about the process, says Josh Marquis, a prosecutor and board member of the National District Attorneys Association, “Someone identifying someone with 100 percent is very powerful piece of evidence in court – though like all eyewitness testimony it has to be tested by a jury.”

Marquis adds that he feels it is was unfair for details about how suspects were identified to be disclosed at the current phase of the case.

The police report also included the accuser’s account of being hit, kicked, and strangled. A nurse’s report based on an examination of the alleged victim on the night of the party reportedly found signs of trauma on her body. She claims to have had bruises on her neck and shoulders.

If the results of the lineup are not admitted as evidence, it could bring the trial to a halt. As lawyer and law professor Rudolf describes it, a central question would become, “Did this photo lineup so taint the witness’ identification that she should not be able to identify people in court? If it did, the case is over.”

Wells, the identification expert, says there were some elements of the identification process that were done right. Showing the pictures in a sequence rather than all at once, side-by-side is the generally preferred method. When shown the pictures all at once, alleged victims tend to compare them and simply choose the person who looks most like an attacker; on the other hand, when shown the pictures one at a time, they are more likely to pick the actual attacker him or herself.