dissenting:
It is now settled that the district court erred in holding that Washington’s counsel was competent. I think the district court also erred in holding that counsel’s incompetence did not prejudice Washington because the results of the semen tests were inconclusive. The results were inconclusive, so the argument runs, because the stains were mixed with vaginal fluid. Prejudice arises from the defense counsel’s failure to show the jury how this inconclusive evidence undermined the Commonwealth’s ease.
This evidence establishes that there were two levels of “inconclusive” evidence. The first is whether the semen was in fact mixed with vaginal fluid. The second is the “inconclusive” result of such a mixture. Defense counsel presented none of this evidence to the jury.
It is undisputed that if the semen stains were not mixed with vaginal fluid they could not have come from Washington or the victim’s husband. The most likely source of the stains was the Commonwealth’s original suspect, James Pendleton. The laboratory report, prepared by a technician working in the Virginia Bureau of Forensic Science shortly after the crime and long before Washington became a suspect, contains no statement about a mixture of semen with vaginal fluid.
The laboratory report unequivocally identifies the semen:
Spermatozoa and/or spermatozoa heads were identified in an extract of each of five (5) other stains. Further test results on four (4) of these stains indicate that the secretions in each are type A, PGM 1. Further test results on the remaining stain indicate that the secretions are type A. The amount of this stain was insufficient for typing in the PGM system.
The report also noted:
Spermatozoa were identified in an extract of the vaginal swabs. These swabs were heavily stained with blood. Further test results on these bloodstained swabs indicate that the type is A, PGM 1.
The report says nothing about vaginal fluid making analysis of the semen on the blanket or on the swabs inconclusive.
Type A, PGM 1 did not match the type of the victim’s husband. But according to the laboratory report it did match James Pendle-ton’s type A, PGM 1. Pendleton was the original suspect. Why he was not prosecuted is not disclosed in the record; this, however, is unimportant. Washington is not obliged to point out the guilty party. Holland v. Commonwealth, 190 Va. 32, 40-41, 55 S.E.2d 437, 441 (1949). What is significant is that when Pendleton was the suspect there was no talk about vaginal fluid masking the semen analysis. Everything was clear. The type could be identified, and it was.
In short, all went well until Washington turned up with blood type O, PGM 2-1. He is a secretor; that is, his blood type 0 would identify his semen. There is nothing in the Commonwealth’s report disclosing the presence of a stain left by a person with type 0, PGM 2-1. Furthermore, there is nothing in the report about a mixture. The victim’s husband, therefore, could not have left the stain, if we take the report at its face value.
The Commonwealth’s laboratory report, together with the Commonwealth’s subsequent test of Washington’s blood when he was arrested about a year later, exonerated *1293Washington. Because Washington’s counsel did not appreciate the significance of the report, he did not present this fact to the jury-
One of the defendant’s experts at the habe-as hearing was Dr. Henry C. Lee, chief criminologist for Connecticut and director of the Connecticut State Police Forensic Laboratory. As a result of his review of the Commonwealth’s report and Washington’s blood type, he testified unequivocally that in his opinion the most probable depositor of the stains was James Pendleton. Moreover, as Dr. Lee testified, if there had been no mixture, the victim’s husband could not have left the stains. Dr. Lee also questioned the existence of a mixture.
To mend its own report, the Commonwealth now advances the theory that vaginal fluid masked the stains. But this theory is not based upon scientific evidence. It is based upon anecdotal testimony of the technician that she assumed vaginal fluid was present because she usually observed it in other instances. Nevertheless, she admitted to having seen pure semen stains unmixed with vaginal fluid in some other cases. Although a small amount of vaginal fluid would not have masked Washington’s type 0, PGM 2-1 deposit, the technician could not say how much vaginal fluid was in the mixture she hypothesized.
This evidence cannot be viewed in isolation. To determine whether defense counsel’s failure to present this evidence prejudiced Washington, it must be viewed in the context of the other evidence — -the shirt attributed to Washington through his admission of ownership and his confession. The shaky evidence about the shirt is fully documented in Washington v. Murray, 952 F.2d 1472, 1478-79 (4th Cir.1991), and there is no need to repeat, it in detail. Suffice it to say that in a thorough, investigative search of the crime scene the shirt was not discovered. The victim’s mother-in-law found it in a dresser drawer that contained clothes belonging to the victim and her husband. A member of the family apparently thought the shirt belonged to him and wore it. The laboratory report shows that hairs in the shirt were consistent with the hair of James Pendleton. When defense counsel requested comparison with Washington’s facial hair, the request was denied. Washington’s sister, who laundered his clothes, testified that the shirt was not his. All of this was presented to the jury.
Because of counsel’s deficiencies, however, the jury were never told that the laboratory report showed that both the stains on the blanket and the hair on the shirt were consistent with James Pendleton’s physical characteristics.
Washington v. Murray recounts the discrepancies in Washington’s confession, and repetition here would be redundant. 952 F.2d at 1478 n. 5. It must be remembered that Washington had an I.Q. in the range of a child of 10.3 years and that he was “easily led.” 952 F.2d at 1475, 1478 n. 5.
It will not do to dismiss the questionable confession because all of the evidence about it was disclosed to the jury. Of course, the confession was admissible, but admissibility is only the first step. Jurors are the ultimate judges of a confession, and this jury was not able to consider the confession in light of the laboratory report, which along with Washington’s blood type, exonerated Washington. It is difficult to assume with confidence that the jury would have accepted the Commonwealth’s present anecdotal theory that the semen type was masked by vaginal fluid. The prosecutor has yet to explain how the stains could have been matched with James Pendleton when the report was written and Pendleton was the suspect but that now the alleged presence of vaginal fluid masks the same stains.
To establish prejudice, “[t]he defendant must show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Washington v. Strickland, 466 U.S. 668, 694, 104 S.Ct. 2052, 2068, 80 L.Ed.2d 674 (1984). Whether a defendant has met this test is ultimately a matter of judgment reached after rigorous scrutiny of the record. I believe that Washington has shown the prejudice required by *1294Strickland. He is entitled to a new trial on the issues of guilt and death. I would grant the writ.