KENNARD, J., Dissenting. In 1980, petitioner opened fire on a group of police officers who were trying to arrest him in a crowded bar, killing one officer, seriously wounding two others, and wounding two of the bar’s patrons. At trial, he asserted a defense of diminished capacity, claiming that he was under the influence of amphetamines when he began shooting and therefore acted without premeditation and deliberation or malice. He was convicted of capital murder and sentenced to death. He then filed a petition for writ of habeas corpus, contending that the prosecution had breached its duty to provide him with material, exculpatory evidence when it failed to disclose to his attorneys that a preliminary screening of a sample of petitioner’s blood taken shortly after his arrest showed that petitioner might have had a different drug, phencyclidine (hereafter PCP), in his bloodstream. A majority of this court agrees with his contention. I do not.
In my view, the evidence at the proceedings on habeas corpus, held well over a decade after the trial when memories of the events in question had faded, is inconclusive as to whether the prosecution failed to provide the defense with the test result in question. Thus, petitioner has not met his “heavy burden” (People v. Duvall (1995) 9 Cal.4th 464, 474 [37 Cal.Rptr.2d 259, 886 P.2d 1252]) of proving that the prosecution violated his right to due process by failing to disclose the result. Far more important, however, it is inconceivable that the result of this highly unreliable preliminary screening could have altered the outcome of petitioner’s trial. There was no evidence that petitioner was under the influence of PCP when he killed Officer Donald Reed, apart from his own self-serving statement that he had used the drug in the past; and he made no pretense of having used it in the days leading up to the shooting. As a result, the test result in question was not “material,” and therefore did not trigger the prosecution’s duty to disclose “material” exculpatory evidence.
I
The testimony at the guilt phase of petitioner’s trial is accurately summarized in this court’s opinion affirming petitioner’s conviction on appeal:
[893]“In June 1980 defendant was a wanted man; he had failed to appear for a jury trial and another criminal hearing, and two bench warrants were issued for his arrest. After telling his former live-in girlfriend he was not going back to jail and did not want to die in prison, defendant bought a gun and changed his name to Gordon Mink.
“Meanwhile the Garden Grove police were looking for him. At an evening prepatrol briefing on Saturday, June 7, the police department distributed a flier to all officers. The flier contained defendant’s name, photograph, and description; it noted there were outstanding warrants for his arrest, described his car, and listed the address where the car was last seen.
“Shortly thereafter, Officer Mclnemy and his partner, Reserve Officer Henninger, found the described car at an apartment complex. They watched it from their marked patrol car for about 45 minutes before they left to handle other pressing calls. While investigating another incident about 11 o’clock later that night, the two officers again noticed defendant’s car—this time at the Cripple Creek Bar.
“They called for assistance, and Garden Grove Officers Reed and Overly quickly arrived. After discussing the flier distributed earlier that evening, all four officers—all in full uniform—entered the crowded bar through two separate doors and worked their way to the center of the room.
“Defendant, who was sitting in the comer with a group of other ‘motorcycle-type people,’ saw the officers enter; a nearby patron heard him [curse and] say ‘the pigs are here,’ as he started for the door. The officers recognized defendant and moved in his direction. At the door, Officer Reed caught up with defendant and put his hand on defendant’s shoulder. Before any of the officers could draw his weapon, defendant pulled a gun and fired at least eight times. Two lethal shots hit Officer Reed; three shots gravely wounded Officer Overly; Officer Henninger was seriously wounded; a private citizen, Terezia, suffered permanent and grave injury after being shot between the eyes; and another citizen, McKinney, was shot in the leg.
“Defendant fled and hid in some bushes outside the bar. About two hours later, with numerous officers at the scene, he was found crouched in the dirt. As he was brought out of the bushes an officer called out, ‘Where’s the gun?’ Defendant stated, T threw it.’ His gun, hat and keys were thereafter found nearby.
“Defendant testified in his defense and presented expert witnesses who suggested he may have suffered from diminished capacity because of drug [894]abuse at the time of the incident. Defendant admitted he had suffered a 1970 felony conviction in Florida for ‘burglary.’ He detailed his extensive drug abuse history and claimed to have been under the influence of methamphetamines on the night in question and that he remembered nothing of the events in question. In rebuttal, the People established that defendant’s blood sample, taken shortly after his arrest, showed no presence of drugs and specifically, the test for methamphetamine was negative. Moreover, numerous officers who dealt with defendant in the four to five hours after his arrest testified that although he ‘stank like a pig’ he behaved normally and did not appear to be under the influence of any drug.” (People v. Brown (1988) 46 Cal.3d 432, 440-441 [250 Cal.Rptr. 604, 758 P.2d 1135].)
At trial, petitioner testified that he had used PCP and many other drugs on previous occasions, but the only drug he claimed to have taken in the days leading up to the shooting was methamphetamine. Dr. Kaushal Sharma, a psychiatrist testifying for the defense, asserted that petitioner’s previous use of PCP could have affected him at the time he killed Officer Reed even though he had not used the drug at or near the time of the shooting. Dr. Sharma explained that this could occur because PCP is stored in the fatty tissues of the body and may reenter the bloodstream long after the user first took the drug, producing “flashbacks.”
According to Dr. Sharma, some people act in a bizarre or violent manner when under the influence of PCP, while others can take the drug without showing any outward symptoms. A PCP user’s behavior during a flashback depends on “what they were acting like when they took the initial PCP”: Users who initially acted in a “strange” manner would generally exhibit similar behavior during a flashback, while those whose actions were normal when initially taking the drug would act similarly during a flashback. Petitioner had told Dr. Sharma that when he took PCP he sometimes had “strange thoughts” or “acted in a strange manner” but he “had not had any bad trips on PCP.”
The jury found petitioner guilty of murdering Officer Reed (Pen. Code, § 187) and found true a special circumstance allegation that he intentionally killed a peace officer engaged in the performance of his duties {id., § 190.2, subd. (a)(7)). At the penalty phase, the jury imposed a sentence of death.
In his petition for writ of habeas corpus, petitioner alleged that law enforcement authorities had performed two tests on his blood sample: (1) a gas chromatography mass spectrometry (GC/MS) test, which detected no drugs and was introduced against petitioner at his trial, and (2) a radioactive immunoassay (RIA) screening of his blood, which was positive for PCP but [895]was never disclosed to the defense. Petitioner contended that by failing to disclose the latter result, the prosecution violated his right to due process of law, which imposes on the prosecution an “affirmative duty to disclose evidence favorable to a defendant,” even in the absence of a specific request. (Kyles v. Whitley (1995) 514 U.S. 419, 432 [115 S.Ct. 1555, 1565, 131 L.Ed.2d 490]; see also United States v. Bagley (1985) 473 U.S. 667, 683 [105 S.Ct. 3375, 3384, 87 L.Ed.2d 481]; Brady v. Maryland (1963) 373 U.S. 83, 87 [83 S.Ct. 1194, 1196-1197, 10 L.Ed.2d 215].) We issued an order to show cause and an order of reference, directing the referee to determine whether the prosecution disclosed the RIA test result to the defense, and whether the GC/MS and RIA test results could be reconciled.
The referee (Judge Daniel J. Didier of the Orange County Superior Court) concluded that the prosecution had disclosed the RIA test result to the defense. He also found that the difference in the GC/MS and RIA results was attributable to the unreliability of the RIA test, which often reported the presence of PCP in blood samples in which no PCP was present.
The majority rejects each of these findings. It concludes that the prosecution failed to provide the RIA test result to the defense, and that the inconsistent test results, while possibly attributable to the limitations of the RIA test, may also have been attributable to other factors that would be consistent with the hypothesis that petitioner was under the influence of PCP when he killed Officer Reed. Finally, it concludes that the undisclosed test result was “material”; that is, the result “could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict” (Kyles v. Whitley, supra, 514 U.S. at p. 435 [115 S.Ct. at p. 1566], fn. omitted), and that therefore petitioner’s conviction and death sentence must be vacated.
The majority is wrong in rejecting the referee’s finding that the prosecution did disclose the RIA test result to the defense. Moreover, although I agree with the majority that there are several possible explanations for the inconsistency between the RIA and GC/MS test results, I conclude that the RIA test was not “material,” because it could not possibly have affected the outcome of petitioner’s trial.
II
The first question we asked the referee to determine was this: “Did the prosecution disclose the positive PCP finding to petitioner, his investigator, or his counsel before or during trial?” The referee found it “very difficult” to give a yes or no answer, because more than a decade had passed since the [896]trial. There was strong circumstantial evidence that the prosecution had disclosed the test result to the defense, but there was equally strong circumstantial evidence that the result had not been disclosed. After weighing the conflicting evidence, the referee found that the prosecution had disclosed the test’s positive PCP result to the defense.
It is undisputed that the Orange County Sheriff-Coroner Department, Forensic Science Services (hereafter the crime lab) performed GC/MS and RIA tests of petitioner’s blood sample, and that the former test was negative for PCP and several other controlled substances while the latter test was positive for PCP. The crime lab prepared a detailed “worksheet” that recorded the results of both tests; the lab’s “result sheet,” however, indicated only that PCP was “not detected” in petitioner’s blood sample. Although no witness expressly explained why the result sheet did not mention the positive RIA test, it appears to have been omitted because crime lab scientists believed that negative RIA test results were reliable but positive results were not, and they used the RIA test as a preliminary screening mechanism to determine whether they needed to perform the more accurate GC/MS test. In 1980, when the testing occurred, the crime lab’s policy in criminal cases was to send copies of the result sheet to the prosecutor and defense counsel, but to send a copy of the worksheet to either attorney only upon request. Disclosure of the worksheet required the approval of a supervising criminalist.
The record before this court contains no evidence that the crime lab furnished the Orange County District Attorney’s office with a copy of the worksheet, which showed the positive RIA test, nor does it indicate that anyone in that office ever learned of the result of the RIA test. Prosecutors, however, have a duty to disclose not only material, exculpatory evidence actually known by the attorney prosecuting the case, but also evidence known to the investigating agencies acting on the prosecutor’s behalf. (Kyles v. Whitley, supra, 514 U.S. at p. 440 [115 S.Ct. at pp. 1568-1569].) Thus, the prosecution in this case was obligated to disclose any material, exculpatory evidence possessed by the crime lab, a county agency that was analyzing the blood tests for the district attorney’s office. The crucial inquiry therefore is whether the crime lab ever disclosed the RIA test result to petitioner or his attorneys.
Deputy Public Defender Michael Beecher (now an Orange County Municipal Court Judge) initially represented petitioner following his arrest on June 8, 1980. In November 1980, petitioner elected to represent himself, and he handled the pretrial proceedings the year thereafter, while Beecher provided assistance as “backup” or “advisory” counsel. In November 1981, the [897]trial court appointed private attorney Daye Shinn to represent petitioner, and Shinn continued to represent petitioner throughout the balance of the proceedings in the trial court. There is no evidence that the crime lab ever provided a copy of the worksheet to petitioner, Defense Attorney Shinn, or Shinn’s investigator, or that anyone ever informed any of them of the worksheet’s existence. Whether the crime lab furnished the worksheet to Attorney Beecher, however, was a hotly contested issue at the reference hearing.
Deputy Public Defender Beecher’s file notes show that on October 16, 1980, he telephoned Mary Graves, a supervising criminalist at the crime lab, and asked for reports on petitioner’s blood testing, and that Graves told him she would send a report. But his file, which was retained by the public defender’s office after it ceased to represent petitioner, contains only the result sheet, not the worksheet on which the PCP-positive RIA test result was recorded.1 Beecher testified at the reference hearing that he did not recall receiving the worksheet, and that his notes made no mention of the worksheet. Had he seen it, he said, he would have regarded the worksheet’s information as “very significant,” would have noted it in his file, and would have consulted with a PCP expert.
Because the reference hearing was held 16 years after the crime lab’s testing of petitioner’s blood sample, it was not surprising that no one at the crime lab had any recollection of sending the worksheet of petitioner’s blood test to Defense Attorney Beecher. But the bottom of the worksheet carried this notation: “Bercher [sic], Pub. Def. B-100 10-16-80.” Frank Fitzpatrick, the chief criminalist in charge of managing the crime lab’s clerical staff in October 1980, testified that this notation meant that on October 16, 1980, the lab sent copies of both the result sheet, which contained only the lab’s ultimate finding that PCP was “not detected,” and the more detailed worksheet, which noted that the preliminary RIA test was positive for PCP, to Attorney Beecher at his office in room B-100 of the Orange County Courthouse.
The referee found the notation on the worksheet, mentioned above, to be persuasive evidence that the prosecution had sent a copy of the worksheet to Beecher, but also found Beecher’s testimony persuasive evidence to the contrary. It resolved this conflict in favor of the prosecution.
In the view of the majority, however, the referee did not find that the prosecution had provided the worksheet to the defense. The majority states [898]that the referee’s findings “do not accurately respond to the specific question posed in our reference order—did the prosecution disclose the RIA result to petitioner or anyone acting on his behalf?” (Maj. opn., ante, at p. -878, original italics.) On the contrary, the referee’s finding is explicit: “The evidence preponderates that ... the positive RIA finding, was . . . disclosed and duly forwarded to defense counsel before trial, as requested.” The only aspect of this finding that could conceivably be viewed as unresponsive is its failure to specify who had disclosed to the defense the worksheet indicating the positive RIA finding. The majority apparently considers this a fatal flaw, pointing out that the prosecuting attorney “had no knowledge of the worksheet and thus could not have disclosed it.” (Ibid.) Thus, the majority concludes, it need not determine whether the crime lab had furnished the test result to the defense, because the “prosecution”—by which the majority presumably means the district attorney’s office—did not do so. (Ibid.)
The majority is wrong. As the majority itself acknowledges, courts have consistently “ ‘decline[d] “to draw a distinction between different agencies under the same government, focusing instead upon the ‘prosecution team’ which includes both investigative and prosecutorial personnel.” ’ ” (Maj. opn., ante, at p. 879, quoting United States v. Auten (5th Cir. 1980) 632 F.2d 478, 481.) Here, the Orange County District Attorney’s office and the Orange County Sheriff-Coroner Department’s crime lab, the agencies involved here, worked on the same side, that of the prosecution. In short, they were the prosecution’s team. Therefore, so long as disclosure of material evidence was made to the defense by any part of that team, whether by a member of the district attorney’s office or a member of the crime lab, the prosecution had done its duty.
The majority apparently also concludes that the referee found that the defense never received the worksheet. (Maj. opn., ante, at p. 878.) Although the referee’s report on this issue is somewhat unclear, in my view he made no such finding. Rather, the referee found that there was “credible” circumstantial evidence that the worksheet had not been disclosed to the defense, but he also found credible circumstantial evidence of such disclosure. The referee resolved the conflict adversely to petitioner, who had the burden of proof.2 [899]The referee’s resolution was proper. As this court has explained: “Because a petition for a writ of habeas corpus seeks to collaterally attack a presumptively final criminal judgment, the petitioner bears a heavy burden initially to plead sufficient grounds for relief, and then later to prove them. ‘For purposes of collateral attack, all presumptions favor the truth, accuracy, and fairness of the conviction and the sentence; defendant thus must undertake the burden of overturning them. Society’s interest in the finality of criminal proceedings so demands, and due process is not thereby offended.’ ” (People v. Duvall, supra, 9 Cal.4th 464, 474, original italics.)
This court gives great weight to a referee’s findings of fact when they are supported by substantial evidence, because the referee had the opportunity to observe the demeanor of witnesses and their manner of testifying. (In re Marquez (1992) 1 Cal.4th 584, 603 [3 Cal.Rptr.2d 727, 822 P.2d 435].) Curiously, in reaching its conclusions, the majority makes no mention of this well-established standard of review. Here, the evidence at the reference hearing was inconclusive. According the referee’s findings the weight they are due, I conclude that petitioner did not meet his heavy burden of establishing that the prosecution failed to disclose to the defense the RIA test result indicating the presence of PCP in petitioner’s blood sample.
IH
The second question we asked the referee to resolve was this: “Did the positive PCP finding from the RIA test indicate that there was PCP or a PCP analog in petitioner’s blood at the time of the crimes? If so, did the subsequent negative PCP finding from the GC/MS test establish that there was no PCP or PCP analog in petitioner’s blood at the time of the crimes? How can the results of the two tests be reconciled?”
The referee found that (1) the RIA test’s positive PCP finding did not indicate that there was PCP or a PCP analog in petitioner’s blood; (2) the [900]subsequent negative finding by the GC/MS test established that there was no PCP or PCP analog in petitioner’s blood; and (3) the conflicting results of the two tests could be “reconciled” by treating the RIA test result as incorrect.
The referee’s finding that the GC/MS test was right and the RIA test was wrong is, according to the majority, but one explanation for the discrepancy between the two tests. There are other explanations for the discrepancy, the majority notes, that leave open the possibility that the RIA test result was correct. I agree with the majority that it is possible, albeit unlikely, that the RIA test result was accurate.
The most persuasive evidence regarding the reliability of the RIA test was offered by Dr. Vina Spiehler, a noted forensic toxicologist whose expert testimony was credited by the referee. Dr. Spiehler is a former president of the Society of Forensic Toxicologists and the California Association of Toxicologists, a former editor of the Toxicology Division Newsletter of the American Academy of Forensic Sciences, and the author of numerous published studies analyzing the reliability of RIA testing for PCP and other illegal drugs. She was also quite familiar with the procedures at the crime lab where petitioner’s blood sample was tested, having worked there from 1981 (the year after the sample was tested) to 1986.
Dr. Spiehler explained that in 1980 the RIA test was used as a preliminary screen because its negative findings were considered reliable, but its positive findings were not. In other words, if the RIA test found no PCP in a blood sample, then the chances were high that the finding was accurate. But if, as in this case, the RIA test indicated that there was PCP in a suspect’s blood, the result was not reliable, thus necessitating further tests to determine whether PCP was actually present.
Based on studies she and other scientists had performed on the accuracy of the RIA test, Dr. Spiehler concluded that if an RIA test of the type used in 1980 produced a positive result for PCP, there was only a 37 percent probability that the tested blood actually contained PCP. Thus, Dr. Spiehler stated, petitioner’s positive RIA test result meant only that there was about one chance in three that petitioner’s blood actually contained PCP.
At the reference hearing, petitioner argued there were two possible explanations for the conflicting RIA and GC/MS test results that would be consistent with his claim that he had PCP or a PCP analog in his bloodstream at the time of the shooting. Although the evidence did not conclusively exclude either of these possibilities, neither of them was very likely, as I shall explain.
[901]Petitioner’s expert, Dr. Ferris Pitts, pointed out that the crime lab had used the GC/MS test to test petitioner’s blood sample only for PCP, not for any PCP analog. A PCP analog, Dr. Pitts explained, is a substance similar but not identical to PCP, which produces the same psychotropic effects as PCP. Such an analog, he testified, might register a positive result on the RIA test, but a negative result on the GC/MS test, which is more precise.3
Dr. Spiehler, the prosecution’s expert, acknowledged that the GC/MS test might not detect a PCP analog. But she added that most such analogs are chemically so similar to PCP that the GC/MS test would indicate their presence, even if it did not register a positive result. After examining petitioner’s GC/MS test, Dr. Spiehler found no indication that such analogs were present in petitioner’s blood sample. Her testimony was corroborated by Dr. Robert Cravey, a former chief forensic toxicologist for the Orange County Sheriff-Coroner’s Department. Petitioner offered no evidence that any PCP analog whose chemical structure was so different from that of PCP that it would not show up on the GC/MS test was in use in Orange County during 1980.4
Petitioner also asserts that the RIA test can detect smaller concentrations of PCP in a blood sample than the GC/MS test. Thus, he argues, he may have had a quantity of PCP in his blood sample great enough to have affected his thinking and to be detected by the RIA test, but too minute to be detected by the GC/MS test.
The GC/MS test that was run on petitioner’s blood sample had the capability of detecting the presence of as little as 25 nanograms per milliliter of PCP in the blood and might have been able to detect as little as 10 nanograms per milliliter. Because certain printouts relating to the GC/MS test run on petitioner’s blood have been lost or destroyed, it is no longer possible to determine whether that test could have detected quantities between 10 and 25 nanograms per milliliter.
On the other hand, the RIA test could, according to the manufacturer, detect 50 to 70 nanograms of PCP per milliliter of blood and thus could not [902]detect concentrations as small as those that could be found using the GC/MS test. But petitioner’s expert, Dr. Pitts, gave uncontradicted testimony that at the time petitioner’s blood sample was tested almost all laboratories were able to refine the RIA test so that it could measure smaller quantities of PCP. Dr. Pitts said that in 1980 the RIA test could “regularly” detect as little as 10 nanograms of PCP per milliliter of blood and on occasion even as little as 1 or 2 nanograms per milliliter. It is therefore possible that petitioner had a very small quantity of PCP in his blood when tested, and that this tiny quantity was detected by the RIA test but not by the GC/MS test.
It is not at all clear, however, that concentrations of PCP too small to be detected by the GC/MS test can nonetheless have a pharmacological effect. The prosecution’s expert, Dr. Spiehler, testified that the American Association of Clinical Chemistry has found that the drug has such an effect only in concentrations greater than 75 nanograms per milliliter of blood. She added that the most “liberal” scientific organizations have taken the position that such effects can be found in concentrations as small as 7 nanograms per milliliter. As I noted earlier, Dr. Pitts, the defense expert, testified that even smaller quantities of PCP can cause a pharmacological effect. The referee, however, found Dr. Spiehler’s testimony to be more credible than that of Dr. Pitts.
In summary, it is possible that petitioner was under the influence of a PCP analog that was detectable by the RIA test, but which was chemically so different from PCP that it would not appear on the GC/MS test. But because there is no evidence that any such analogs were available in Orange County in 1980, the likelihood of this possibility is slight. Similarly, it is possible that the RIA test that was conducted here may have been capable of detecting concentrations of PCP in petitioner’s bloodstream smaller than those measured by the GC/MS test, and it is possible that those small concentrations could have had a pharmacological effect on petitioner at the time of the shooting. But the likelihood that petitioner’s blood sample had a quantity of PCP too small to be detectable by the GC/MS test, yet large enough to significantly affect his behavior, is not great. Far more likely is the referee’s conclusion that the RIA test’s positive result for PCP was simply an inaccurate “false positive,” which Dr. Spiehler found to be true of most PCP-positive RIA test results. These matters are significant to the resolution of the final question this court must address: whether any failure by the prosecution to disclose the result of the RIA test was material.
IV
A prosecutor’s failure to disclose exculpatory evidence to a criminal defendant before trial does not by itself require that the defendant’s conviction be overturned. Reversal is required only if the undisclosed exculpatory [903]evidence was “material.” (Kyles v. Whitley, supra, 514 U.S. at p. 434 [115 S.Ct. at p. 1566].) Evidence is “material” if there is a “reasonable probability” that the outcome of the trial would have been different had the evidence been disclosed, which occurs when the undisclosed evidence “could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict.” (Id. at pp. 434-435 [115 S.Ct. at p. 1565-1566], fn. omitted.) If the undisclosed evidence is “material,” the defendant’s conviction must be vacated without a separate harmless error review, because the prejudice determination is subsumed within the definition of the term “material.” (Id. at pp. 435-436 [115 S.Ct. at pp. 1566-1567].)
The majority concludes that the prosecution’s alleged failure to disclose the result of the RIA test was “material” within this definition, requiring that his murder conviction be vacated. I disagree.
If the defense had attempted to introduce evidence of the positive RIA test result, it is unclear whether the test result would have been admissible. To justify the test result’s admission, petitioner would have to satisfy the long-established requirements for the admission of scientific tests commonly known as the Kelly/Frye rule. (People v. Kelly (1976) 17 Cal.3d 24 [130 Cal.Rptr. 144, 549 P.2d 1240]; Frye v. United States (D.C. Cir. 1923) 293 F. 1013 [54 App.D.C. 46, 34 A.L.R. 145].) In People v. Diaz (1992) 3 Cal.4th 495, 526 [11 Cal.Rptr.2d 353, 834 P.2d 1171], this court summarized this rule as follows: “Under the Kelly/Frye rule, evidence based on a new scientific method of proof must satisfy three requirements before it may be admitted. First, the party offering the evidence must show that the technique is ‘ “sufficiently established to have gained general acceptance in the particular field in which it belongs.” ’ [Citations.] Second, the proponent of the evidence must establish that ‘the witness furnishing such testimony’ is ‘properly qualified as an expert to give [such] an opinion . . . .’ [Citation.] Third, the proponent must demonstrate that ‘correct scientific procedures were used in the particular case.’ [Citations.]”
It is uncertain whether petitioner could have shown that a PCP-positive RIA test result, as occurred here, was generally accepted by the scientific community as an accurate indicator of the presence of PCP. Although several published decisions have upheld the admission of the results of RIA analysis of human hair as evidence of cocaine use (see State Emp. Sec. Dept. v. Holmes (1996) 112 Nev. 275 [914 P.2d 611]; Matter of Adoption of Baby Boy L. (N.Y.Fam.Ct. 1993) 157 Misc.2d 353 [596 N.Y.S.2d 997]), to my knowledge, no published decision has ever held that the results of RIA blood tests (either positive or negative) for any drug are sufficiently accepted by the scientific community to be admissible in court. According to the prosecution’s expert, Dr. Spiehler, the Food and Drug Administration requires all [904]manufacturers of RIA testing kits to state that a positive finding must be confirmed by a GC/MS test before being reported, and this requirement is echoed in the guidelines of the American Academy of Forensic Scientists and the Society of Forensic Toxicologists. Thus, the scientific community apparently does not view an unconfirmed positive RIA test as a reliable indicator of drug intoxication.5
Assuming for the sake of argument that petitioner could have successfully introduced the RIA test result into evidence, he would have had to overcome an even more formidable hurdle: persuading the jury that (1) he was under the influence of PCP when he killed Officer Reed, and (2) that as a result of the PCP intoxication he lacked premeditation and deliberation or malice. In 1982, when petitioner’s trial took place, the task of convincing a jury of the merits of a diminished capacity claim was difficult in the best of circumstances; as one court explained, “juries take a notoriously dim view of the diminished capacity concept although appellate courts seem to love to explore it in the most erudite manner . . . .” (People v. Huffman (1977) 71 Cal.App.3d 63, 76 [139 Cal.Rptr. 264].)6 Here, there was overwhelming evidence showing either that PCP was not in petitioner’s bloodstream when he killed Officer Reed, or that if it was, it did not significantly affect petitioner’s mental state at the time of the killing. Given this evidence, it is highly unlikely that petitioner would have been successful in persuading the jury that he suffered from diminished capacity caused by PCP intoxication.
Petitioner offered no evidence at trial that his behavior before and after he killed Officer Reed was bizarre or irrational; to the contrary, his conduct [905]appeared goal-directed and rational. Petitioner knew there was a warrant for his arrest; he had changed his name, bought a gun, and told his girlfriend that he did not want to return to prison. When police officers entered the bar on the night of the killing, petitioner said, “Fuck, the pigs are here,” and tried to leave. When one of the officers attempted to detain him, petitioner opened fire, killing one officer and wounding two others. He had sufficient presence of mind to flee, to hide in the bushes, and to dispose of the murder weapon and a hat that might have helped to identify him. When arrested two hours later, he behaved normally and gave no .indication that he was under the influence of any drug. The pupils of his eyes were dilated, which is a symptom of methamphetamine use. PCP, by contrast, causes nystagmus (rapidly bouncing pupils) and causes the pupils to contract. (People v. Bonillas (1989) 48 Cal.3d 757, 782 [257 Cal.Rptr. 895, 771 P.2d 844]; People v. Dunkel (1977) 71 Cal.App.3d 928, 932-933 [139 Cal.Rptr. 685].) There is no evidence that petitioner displayed these PCP symptoms. Moreover, petitioner testified that he used methamphetamine, not PCP, on the day he killed Officer Reed.
The majority speculates that petitioner may have had a PCP flashback when he killed Officer Reed. It points out that at petitioner’s trial a forensic psychiatrist, Dr. Kaushal Sharma, testified that PCP becomes attached to the fatty tissue in the user’s brain and may be released into the user’s system long -after it was originally ingested, causing such flashbacks to occur. Thus, the majority asserts, petitioner’s use of PCP long before he killed Officer Reed could have caused him to be under the influence of PCP at the time of the killing. But Dr. Sharma testified that persons having such flashbacks ordinarily behave in the same manner in which they acted when they initially took the drug. Because petitioner admitted to Dr. Sharma that he “had not had any bad trips on PCP,” it is unlikely that a PCP flashback would have caused his violent behavior at the time of the shooting. Furthermore, there is no evidence that petitioner’s behavior before the officers entered the Cripple Creek Bar was unusual or violent, and the possibility that he conveniently had a flashback at the precise moment they entered the bar is farfetched.
In short, the most reliable scientific evidence (the GCZMS test) showed that petitioner was not under the influence of PCP when he shot and killed Officer Reed, his behavior at the time of the killing was not suggestive of PCP intoxication, and in his own testimony he made no claim of using PCP in the hours or days immediately preceding the shooting. Given this evidence, the RIA test result showing that PCP was possibly present in petitioner’s blood sample could not have altered the outcome of his trial. It therefore did not fall within the scope of the prosecution’s duty to disclose material evidence.
[906]Conclusion
For the reasons set forth above, I would deny the petition for writ of habeas corpus.
Baxter, J., and Chin, J., concurred.
Repondent’s petition for a rehearing was denied May 20, 1998. George, C. J., did not participate therein. Kennard, J., Baxter, J., and Chin, J., were of the opinion that the petition should be granted.
Witnesses testified that Beecher kept meticulous records, making it unlikely that he would have misplaced the test result outside of the file.
This is what the referee said on this issue: “Question one [‘Did the prosecution disclose the Positive PCP finding to petitioner, his investigator, or his counsel before or during trial?’] is very difficult to answer yes or no. A reasonable inference can be drawn, from both the testimony, and the documentary evidence, that copies of both the final results (ex. 1) and the worksheet (ex. 2) were duly forwarded to Public Defender Beecher, as requested. Procedures for obtaining a worksheet were for a requesting attorney to talk to a supervisor, which [899]Beecher did. Notations were to be placed on the documents as to destination, which was done here. The evidence is also credible that Beecher, as well as his successor counsel, were unaware of the existence of the worksheet, thus a reasonable inference that it was not received. The evidence is credible that the District Attorney’s office was not aware of the worksheet, or of the positive RIA finding, and that if any exculpatory evidence had been available, either by the RIA worksheet, or otherwise, that it would have, as a matter of course, been released to defense counsel. There was no evidence that exhibit 2, or any other documentation, was concealed or withheld from either the district attorney or defense attorneys, by the crime lab, or that any police agency or the District Attorney withheld or concealed information. While testimony as to all witnesses on this issue is deemed credible, the evidence preponderates that the documentation, including the positive RIA finding, was freely available, disclosed and duly forwarded to defense counsel before trial, as requested. The answer to question one is yes.”
The GC/MS test for PCP examines only three “channels” on the scale for evidence of the drug. If PCP is present, it will appear on all three channels. Evidence of other drugs chemically related to PCP is likely to appear on one of the three channels, but not all three.
Petitioner’s expert, Dr. Ferris Pitts, testified that a PCP analog, “4-methyl PCP,” is “found on the street and used by individuals.” He said that evidence of this analog would appear on a GC/MS test for PCP, because 4-methyl PCP, like PCP itself, would have a “mass point” at about 200 on the scale used to measure the results of the GC/MS test. On direct examination, he claimed that petitioner’s GC/MS test appeared to show a mass point at about 200. On cross-examination, however, he admitted that he saw only “a little bit of’ a mass point, and that he “would have to do a lot more work to be sure what it was.” The prosecution’s experts, Dr. Spiehler and Dr. Cravey, testified that there was no such mass point in the results of petitioner’s GC/MS test.
I disagree with the majority’s assertion, which is not supported by any citation of authority, that the Attorney General “waived” the argument that the RIA test results would have been inadmissible. (Maj. opn., ante, at p. 890, fn. 12.) Petitioner has the burden of demonstrating a reasonable probability that disclosure of the RIA test would have altered the outcome of his trial, and the Attorney General has properly asserted that petitioner did not satisfy that requirement. Contrary to the majority’s assertion, the Attorney General’s failure to mention the Kelly/Frye test in explaining why that burden has not been met does not prevent this court from relying on the test as a basis for denying the petition for habeas corpus.
The majority also contends that whether the RIA test would have been admissible is “largely beside the point” because the defense could have had the blood sample retested for PCP and “possibly” obtained admissible confirmation of the positive result with a GC/MS test. (Maj. opn., ante, at p. 890, fn. 12.) There is no evidence whatever in the record to support the majority’s speculation that such a test would have verified the RIA test result. Moreover, the record reflects that the prosecution provided the defense with a portion of the blood sample for testing before trial, thus, the prosecution’s alleged failure to disclose the RIA test result did not deny petitioner the opportunity to conduct such a test.
In lune 1982, two weeks after petitioner’s trial, California voters overwhelmingly approved Proposition 8, a voter initiative that, among other things, abolished the diminished capacity defense. (See People v. Saille (1991) 54 Cal.3d 1103, 1112 [2 Cal.Rptr.2d 364, 820 P.2d 588].)