The jury began deliberations late yesterday afternoon in the trial of James Allen Selby, a man suspected of attacking and raping several women near the UA.

The jury is considering 34 counts against Selby. If convicted, he is likely to spend the rest of his life behind bars.

Charges against Selby, 37, include first degree murder, rape, kidnapping and burglary. Selby is accused of attacking five women and a 13-year-old girl in Tucson between October 2001 and May 2002, one of whom was a UA student.

The trial began yesterday with Pima County Assistant Public Defender Michael Mussman calling Mary Walkinshaw to the stand for cross-examination. Walkinshaw, a former criminalist from the Tucson Police Department, handled several of the DNA analyses for the Selby investigation.

Mussman asked Walkinshaw about several minor mistakes he found in paperwork she submitted for the DNA reports.

In one report, Mussman pressed the criminalist to justify the date of the report. In another report, he asked whether a DNA sample had been cleaned three or five times.

Mussman then asked if the former criminalist might have made other mistakes.

Walkinshaw defended her work, saying the errors were minor and not deliberate, adding she corrected the mistakes on the paperwork by writing over the mistakes.

"I'm highly criticized for being anal retentive," Walkinshaw said.

Walkinshaw was followed by Nora Rankin, a senior TPD criminalist, on the witness stand.

Deputy Pima County Attorney Brad Roach asked Rankin about the how the DNA evidence helped determine whether Selby was "the midtown rapist."

"In a random match probability, Mr. Selby is the source of the semen in three cases," Rankin said.

Roach asked about the probability of a random match being identical to Selby's DNA.

Rankin replied the odds were one in 1.4 quintillion. When asked how big a quintillion was, Rankin said it was one followed by 17 zeros. According to Webster's New World Dictionary, a quintillion is defined as a one followed by 18 zeros.

When Mussman cross-examined Rankin, he focused his questions on evidence in one case where there was not an exact DNA match.

Only one of the DNA tests could not positively identify Selby, Mussman said.

Rankin testified because the sample had DNA from at least two people and, it was impossible from the sample collected to conclude whether the DNA came from Selby. The sample, coming from a hand scrub performed at the crime scene, included the victim's DNA profile and may have also included Selby's DNA.

Mussman concluded the results could not be used to convict Selby.

"It could be someone else other than Mr. Selby," Mussman said.

Rankin countered the odds of it being someone else's DNA was one in 280 million.

The closing arguments concluded yesterday afternoon, and the jury has begun deliberations.

Selby was recently convicted in Colorado for raping a 56-year-old woman and was sentenced to 20 years in prison there.

Selby is suspected in three other states of raping or attempting to rape several other women.