Opinion

Defendant appeals from a judgment of conviction entered on a jury verdict finding him guilty of lewd and lascivious conduct upon a child under 14 years of age (Pen. Code, § 288) while armed with a firearm (Pen. Code, § 12022).

The sole issue at trial was the identification of the defendant as the perpetrator of the crime. The facts of the crime and the somewhat convoluted identification process are as follows. In the twilight hours of February 14, 1976, Barbara, age 12, and 2 girlfriends, Lou and Therese, made a purchase at a doughnut shop. An employee of the shop later testified she noticed two men, also customers, watching the girls. She observed that one of the men had long dirty red hair and a beard, and had a gun in his pocket.

The three girls then walked over to the neighborhood school grounds to sit and talk. As they entered the grounds, they were approached by a white man whom they had earlier observed at the doughnut shop. He asked the girls if they would do him a favor, then produced a gun and ordered the girls to sit down. Lou began to cry and the other two girls tried to calm her. The man instructed Therese to remain with Lou and directed Barbara to follow him into some nearby bushes where he attempted to rape her; the victim did not confirm a penetration. The victim later testified that it was getting dark at this time and what she remembered most about the assailant was his long reddish brown hair and beard.

The victim reported the attack to the police and was examined by a police doctor later that same evening. The examination confirmed the presence of semen in the region of the victim’s vagina. The semen sample was not tested to identify the donor class of the rapist; it was retained by the police on a smear slide but no further measures such as refrigeration were taken to assure its preservation for future identification analysis. Pursuant to discovery defense counsel later acquired the slide. Not until after trial, however, did a belated laboratory analysis reveal some type B blood group activity on the slide. A more extensive analysis could not be performed because the slide had not been properly preserved. As no blood sample was taken from the victim, it was never determined whether the observed blood type was that of the victim or the attacker.

Two weeks after the event, the three girls went to the police station to attempt to identify the attacker from police photographs. One of the girls, Lou, selected a mug shot of the defendant; she informed the other two girls she had found the assailant, and, after some discussion, the other girls agreed. The police officer gave the girls the defendant’s mug shot to take home to show two other possible witnesses, one being the mother of one of the girls, Mrs. S. She had not been a witness to the crime, but had reported that on the evening of the attack a man had made a lewd remark to her in the vicinity of where the rape occurred later that night. When the girls showed Mrs. S. the mug shot they had picked out, she identified the photograph as depicting the man who had made the comment to her in the street.

The doughnut shop employee who had observed the armed man watching the girls on the night of the crime reported to the police that the same man had returned a few days later with his hair cut. She later testified that the defendant was definitely not this man.

Two months later Lou, Mrs. S., and the victim returned to the police station for further photographic identification. The police officer showed them eight photographs, one of which was the same photograph of defendant that the girls had in their possession for a week and that Mrs. S. had previously identified. The victim selected two photographs from this group, the defendant and another, but discarded the latter after Lou selected the defendant’s picture.

In early June a police lineup was conducted at defendant’s request. Each person in the lineup was asked to say: “Hold it. I’ve got a gun.” The witnesses were properly instructed not to talk to one another during the lineup. The victim and her two girlfriends all selected the person whose position in the lineup was number one, but he was not the defendant. Mrs. S. selected the defendant, whose position was number three: thus hers was the only corporeal identification of the defendant in the case. After the lineup the girls were informed by the police that they had selected the “wrong” man and Mrs. S. the “right” man.

At trial, the witnesses testified to the above pretrial identifications. The girls also identified in court the same photograph of the defendant they had originally selected as depicting the perpetrator of the crime. The girls testified that they could not identify the defendánt in person, but attributed this to the fact that he had altered his appearance by changing his hair and shaving.

Defendant raises two principal contentions on appeal. The first involves the duty of the prosecution to preserve material evidence. The defendant contends the prosecution had a duty to preserve the assailant’s semen sample recovered from the victim so that the defense could attempt to prove by chemical analysis that the discharge could not have come from the defendant. The second contention is that the failure of defense counsel to object to the introduction of identification testimony resulting from obviously questionable pretrial procedures deprived the defendant of his constitutional right to effective assistance of counsel.

We conclude that if the state recovers a semen sample of one who has made a sexual assault, it has a duty to take reasonable steps to preserve that evidence and to make it available to the defense. However, in the circumstances of this case we hold the defendant has not demonstrated that the people failed in their responsibility to adequately preserve this evidence. On the second issue we hold that the defendant was denied effective assistance of counsel.

I

We first consider whether the prosecution’s failure to adequately preserve the sample of the attempted rapist’s semen deprived the defendant of a fair trial. It is clear that the Constitution does not require the prosecution to make a complete and detailed accounting to the defendant of all police investigatory work on a case. (Moore v. Illinois (1972) 408 U.S. 786, 795 [ 33 L.Ed.2d 706, 713 , 92 S.Ct. 2562 ].) Yet it is well established that the suppression by the state of evidence favorable to an accused, after a request therefor, violates due process, irrespective of the good faith of the prosecution. (Brady v. Maryland (1963) 373 U.S. 83, 87 [ 10 L.Ed.2d 215, 218 , 83 S.Ct. 1194 ].) This court has recognized the prosecutor’s duty to disclose such material evidence favorable to the accused even in the absence of a request from the defense. (In re Ferguson (1971) 5 Cal.3d 525, 532 [ 96 Cal.Rptr. 594 , 487 P.2d 1234 ].) In People v. Hitch (1974) 12 Cal.3d 641, 650 [ 117 Cal.Rptr. 9 , 527 P.2d 361 ], we held that the obligation to disclose the existence of material evidence places on the state a correlative duty to preserve such evidence even without a request therefor, and directed that in the future law enforcement agencies take reasonable measures to ensure its adequate preservation.

The People seek to distinguish Hitch on factual grounds. There, the defendant was convicted of driving under the influence of intoxicating liquor. The results of the breathalyzer test introduced by the prosecution were presumptive evidence of guilt. (Veh. Code, § 23126.) The accuracy of that test was dependent on the precise amount of testing solution used in the test ampoule. ( 12 Cal.3d at p. 644 .) Thus if the ampoule had been preserved, it might have provided a source of evidence to impeach the incriminating test results. In the present case, the People argue that the semen sample was not “material” evidence because (1) the defendant was charged with lewd conduct rather than rape, and hence the presence of semen was not a necessary element of the crime; and (2) the finding of semen merely confirmed the victim’s testimony that an attack occurred, and was not used to identify the defendant.

As in Hitch , we are not in a position to examine the suppressed evidence to decide whether or not it is material. However, evidence lost to the defense because of its destruction by the authorities will be deemed material for the purpose of triggering the due process concerns of Hitch if there is a reasonable possibility that it would be favorable to the defendant on the issue of guilt or innocence. ( 12 Cal.3d at p. 649 .) Contrary to the prosecution’s contention, the rationale of Hitch is thus not limited to circumstances in which the destroyed evidence proves a necessary element of the crime.

At the trial, the police doctor who identified the existence of semen in the victim’s vaginal smear testified that it might have been possible to determine from whom the semen had come. While there are many possible analyses that may be performed on semen to identify the donor, and by corollary, to eliminate others from the class of possible donors, the two analyses deemed most commonly feasible are ABO blood typing and identification of the genetic marker phosphoglucomutase (PGM). (Blake & Sensabaugh, Genetic Markers in Human Semen II. Quantitation of Polymorphic Proteins (1978) 23 J. For. Sci. 717, 727.) A recent discrimination probability study of white males in California indicates that an analysis of ABO type and PGM would eliminate approximately 80 percent of such males as the donor of a particular semen sample.

Thus, an analysis of the semen sample in the present case might have not only impeached the credibility of the prosecution’s witnesses (cf. People v. Ruthford (1975) 14 Cal.3d 399, 407-408 [ 121 Cal.Rptr. 261 , 534 P.2d 1341 ]), but also might have completely exonerated the defendant. Whether or not the police deem the crime sufficiently important to warrant such an identification analysis, they cannot make this decision for the defendant. Accordingly, when a woman has been the victim of an attempted or actual rape and the police recover a semen sample of the assailant, the authorities must take reasonable measures to adequately preserve this evidence.

Such a rule protects not only the due process rights of the defendant, but also society’s interest in the integrity of the judicial system. The duty to preserve critical evidence enhances the reliability of the trial process: if an accused is convicted of rape when available evidence would have exonerated him, not only is he unjustly incarcerated but the actual rapist remains at large. “Law enforcement has failed in its primary function, and has left society unprotected from the depredations of an active criminal.” (Manson v. Brathwaite (1977) 432 U.S. 98, 127 [ 53 L.Ed.2d 140, 162 , 97 S.Ct. 2243 ] (dis. opn. of Marshall, J.).) The duty of the prosecution is not simply to obtain convictions, but to “‘fully and fairly present to the court the evidence material to the charge.’” (People v. Ruthford, supra, 14 Cal.3d 399, 405 .)

In the present case, while the state did retain the semen sample, it failed to meet its burden of establishing that it had “undertake[n] reasonable efforts to preserve the material evidence. ... ” (Italics added.) (People v. Hitch, supra, 12 Cal.3d at p. 650 .) However, under the particular circumstances of the case, the conviction may not be reversed on this ground.

As noted above, the state delivered the semen sample to the defense in response to a pretrial discovery request. At that time defense counsel neither submitted the slide to a laboratory for analysis nor took any measures to assure its preservation. Instead, it appears from the record that counsel gave the slide to the defendant’s father, and only after the defendant was tried and convicted was the semen sample retrieved and sent to a laboratory for analysis. Even at this late date the laboratory was still able to identify the blood type of the sample. While it is theoretically possible that a more complete analysis might have been successfully undertaken had the state refrigerated the evidence, under these circumstances we cannot say that the deterioration of the sample was caused by the state’s inaction. Certainly the prosecution had no duty to preserve the evidence once it was in the hands of the defense.

II

The defendant’s second contention is that he was denied effective assistance of counsel because his trial attorney did not object to the prosecution’s introduction into evidence of impermissibly suggestive pretrial identifications. These identifications were the primary evidence linking him to the crime. The defendant claims that had his counsel made the obvious objections, the pretrial identification evidence might have been suppressed.

The right to effective assistance of counsel was first articulated by the United States Supreme Court in reversing the rape convictions of the “Scottsboro Boys.” (Powell v. Alabama (1932) 287 U.S. 45, 71 [ 77 L.Ed. 158, 171 , 53 S.Ct. 55 , 84 A.L.R. 527 ].) A significant line of recent cases has expanded the constitutional guarantee of an accused’s right to counsel (Gideon v. Wainwright (1963) 372 U.S. 335, 344 [ 9 L.Ed.2d 799, 805 , 83 S.Ct. 792 , 93 A.L.R.2d 733 ]) to include the right to the effective assistance of reasonably competent counsel. (McMann v. Richardson (1970) 397 U.S. 759, 771 [ 25 L.Ed.2d 763, 773 , 90 S.Ct. 1441 ]; United States v. DeCoster (D.C.Cir. 1973) 487 F.2d 1197, 1202 ; People v. Pope (1979) 23 Cal.3d 412, 423-424 [ 152 Cal.Rptr. 732 , 590 P.2d 859 ].) The right to adequate assistance of counsel, grounded on the Sixth Amendment of the United States Constitution and section 15 of article I of the California Constitution, is in addition to the general due process protection of a fair trial; it focuses “on the quality of the representation provided the accused.” (People v. Pope, supra, 23 Cal.3d at p. 423 .)

In Pope a majority of the court articulated the analysis involved in a determination of whether a defendant has been denied adequate assistance of counsel. The appellant has the burden of showing “that trial counsel failed to act in a manner to be expected of reasonably competent attorneys acting as diligent advocates. In addition, appellant must establish that counsel’s acts or omissions resulted in the withdrawal of a potentially meritorious defense.” (Id. at p. 425.) The Pope standard is applicable to retained counsel. (People v. Frierson (1979) 25 Cal.3d 142, 161-162 [ 158 Cal.Rptr. 281 , 599 P.2d 587 ].)

Turning to the case at bar, we note intially that a penetraing concern as to the propriety of a pretial identification should be a commonplace consideration to any attorney engaged in criminal trials. “In any case in which the defendant has been identified at a pretrial confrontation conducted without counsel, defense counsel.. . should consider an attack upon the identification procedure. . .on the ground that the confrontation was unfair.” (Criminal Defense Techniques (Hall & Eisenstein edits. 1979) § 2.01 [2], p. 2-8.) The record of the instant case strikingly reveals, from the first day of the trial, that the sole issue was the validity of the pretrial identification of the defendant as the assailant. During voir dire, every prospective juror was asked by the prosecutor whether he or she could vote for a conviction when the only evidence linking the accused to the crime was mere photographic identification. In addition, there was no in-court corporeal identification of the defendant by any witness. In these circumstances, if an objection to the identification evidence would have been potentially meritorious, and thus should have been argued to the trial court, counsel’s failure to so object denied defendant a trial on the key issue of the case.

Since an objection to the identification evidence would have been adjudicated outside the presence of the jury, there could be no satisfactory tactical reason for not making a potentially meritorious objection. (People v. Pope, supra, at p. 426 .)

In order to demonstrate that the alleged incompetency of his trial counsel in not objecting to the identification evidence denied him a potentially meritorious defense, the defendant must present a convincing argument that the pretrial identification procedure “resulted in such unfairness that it infringed his right to due process of law.” (Stovall v. Denno (1967) 388 U.S. 293, 299 [ 18 L.Ed.2d 1199, 1205 , 87 S.Ct. 1967 ]; People v. Caruso (1968) 68 Cal.2d 183, 184 [ 65 Cal.Rptr. 336 , 436 P.2d 336 ].) Our task is thus to assess the facts and circumstances of the identifications to determine whether they were “so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.” (Simmons v. United States (1968) 390 U.S. 377, 384 [ 19 L.Ed.2d 1247, 1253 , 88 S.Ct. 967 ]; People v. Blair (1979) 25 Cal.3d 640, 659 [ 159 Cal.Rptr. 818 , 602 P.2d 818 ].)

We have carefully reviewed the critical facts relating to identification. As previously noted, about two weeks after the crime the victim and her two girlfriends went to the police station to attempt to identify a suspect. No effort was made to separate the witnesses so as to assure independent appraisals. The defendant’s mug shot was initially selected by Lou, and she told the other two girls that she had found the assailant. At that time, Barbara, the victim, was considering another suspect’s photograph. Barbara testified to what happened next: “I looked at the hair and mustache and stuff, and the color of the hair was a reddish brown, and he sorta looked like the guy. And the other [photograph] didn’t look nothing hardly even like him. So we all agreed it was him.”

The foregoing testimony suggests the witnesses felt constrained to select one of the mug shots as the assailant, and the defendant’s photograph initially selected by one of the girls was then collectively declared to be the one most resembling the assailant. We do not doubt that it may be helpful for the police to determine which of their suspects most resembles an assailant, but this procedure does not address the ultimate question, i.e., whether the defendant is in fact the assailant. Furthermore, it appears the identification here was a product of “mutual reinforcement of opinion” among the witnesses (Clemons v. United States (D.C.Cir. 1968) 408 F.2d 1230, 1241 , 1245 & fn. 16), and it is unclear from the record whether or not the girls could have independently identified the defendant. It is clear they did not do so. Furthermore, they were unable to do so at the ensuing lineup.

Once the defendant’s mug shot had been selected in this fashion, the police officer gave the photograph to the girls to take home with them to show Lou’s mother, Mrs. S. Although the record is ambiguous, it appears the girls retained the defendant’s photograph at home for at least a week. When the girls showed Mrs. S. the single mug shot they had selected as the assailant, she agreed it was the same man who had made a lewd remark to her on the night of the crime.

It would be difficult to conceive a more impermissibly suggestive identification procedure. The mere showing of suspects singly to a witness for identification has been widely condemned. (Stovall v. Denno, supra, 388 U.S. 293, 302 [ 18 L.Ed.2d 1199, 1206 ]; Foster v. California (1969) 394 U.S. 440, 443 [ 22 L.Ed.2d 402, 406 , 89 S.Ct. 1127 ].) Here, Mrs. S. was effectively told, “this is the man who molested your daughter’s friend.” The danger of error in identification is at its greatest when the police display only the picture of a single individual and it is heightened when the witness has indications that there is other evidence that the person in the photograph committed the crime. (Simmons v. United States, supra, 390 U.S. 377, 383 [ 19 L.Ed.2d 1247, 1252 ].) The extraordinary suggestiveness of this identification procedure raises considerable doubt that the prosecution would have been able to introduce Mrs. S.’s identification over a timely objection by defense counsel.

We next consider whether Mrs. S.’s subsequent identification of the defendant at a lineup was also objectionable. The question is whether the identification she made at the properly conducted lineup resulted from her prior viewing of defendant’s photograph “or instead [from] means sufficiently distinguishable to be purged of the primary taint.” (People v. Martin (1970) 2 Cal.3d 822, 831 . [ 87 Cal.Rptr. 709 , 471 P.2d 29 ]; Wong Sun v. United States (1963) 371 U.S. 471, 488 [ 9 L.Ed.2d 441, 455 , 83 S.Ct. 407 ].) The danger is that the witness, having seen a suggestively displayed picture, will “retain in [her] memory the image of the photograph rather than of the person actually seen. ...” (Simmons v. United States, supra, 390 U.S. 377, 383-384 [ 19 L.Ed.2d 1247, 1253 ].) If trial counsel had made the appropriate objection, the prosecution would have had the burden of establishing by clear and convincing evidence that Mrs. S.’s lineup identification was purged of the taint of the prior illegal procedure. (People v. Caruso, supra, 68 Cal.2d 183, 189-190 .)

It is at least questionable whether the prosecution could have met this burden. The mere fact that Mrs. S. testified at trial that her identifications stemmed from her observation at the time and place of the street encounter begs the critical inquiry, i.e., “How did her testimony as to her specific observations tend to show that her in-court identification was not infected with the taint of the illegal pretrial confrontation?” (People v. Martin, supra, 2 Cal.3d 822, 833.) By her own admission, Mrs. S. had only seconds to glance at a passerby who made a crude remark to her; compared to that miniscule time span, she had more recent access of at least a week to the defendant’s photograph.

For the foregoing reasons we conclude that trial counsel’s failure to obtain an adjudication of the admissibility of the critical identification evidence against his client deprived the defendant of constitutionally adequate assistance. The “trial” that might have determined the defendant’s fate could very well have taken place not in the courtroom but in the illegally suggestive pretrial identification procedures. (Cf. United States v. Wade (1967) 388 U.S. 218, 235-236 [18 L.Ed.2d. 1149, 1162, 87 S.Ct. 1926 ].)

The judgment is reversed.

Bird, C. J., Tobriner, J., Clark, J., Richardson, J., Manuel, J., and Newman, J., concurred.

Respondent’s petition for a rehearing was denied February 14, 1980.

The present case is typical of the problem covered by Hitch , in that defendant here was not charged at the time the police physician obtained the semen sample. If a request were a condition to the duty to preserve, the duty might not arise until it became impossible of performance. Grunbaum et al., Frequency Distribution and Discrimination Probability of Twelve Protein Genetic Variants in Human Blood as Functions of Race, Sex, and Age (1978) 23 J. For. Sci. 577, 585, table 10. • Since the genetic systems are statistically independent (id. at p. 583), the probability of two white males having the same ABO blood type and the same PGM type is not the sum but the product of the two individual probabilities. We do not decide the issue on this appeal, and thus do not foreclose the prosecution from attempting to meet its burden upon retrial. In view of the foregoing incidents of inadequacy of representation, we need not reach defendant’s additional claims of his counsel’s incompetence.