Opinion
POCHÉ, J.An information charged Aldon Ray Newsome (defendant) with rape (Pen. Code, § 261, subds. (2) & (3)),1 kidnaping (§ 207) and forcible oral copulation (§ 288a, subd. (d)). He pleaded not guilty. Trial by júry resulted in a deadlock. Newsome thereafter moved to dismiss the charges for failure of the prosecution to preserve a semen sample. The trial court granted the motion and dismissed the information. The People appeal.2 We reverse.
Facts
The record of the motion to dismiss and the referee’s hearing and findings supports the recitals of the evidence captioned below.
Chronology
On April 19, 1979, M. complained of rape. A sheriffs deputy accompanied her to a medical facility for a vaginal.examination.
The parties stipulated there was a lapse of approximately 15½ hours between the time of the alleged rape and the collection of the specimen3 and that within an hour of receiving the sample Deputy Sheriff Sargent of the Santa Clara County Sheriff’s Department refrigerated the sample. The deputy took custody of the swab immediately.
Sergeant Bush removed the specimen from the refrigerator in the medical section of the jail and took it to the criminalistics laboratory on April 23, 1979. The criminalist, Mr. Gadd, air-dried the swab from April 23 to April 25, 1979. On April 25, 1979, Gadd processed the swab to determine if seminal material was present. He obtained a positive presumptive test in-*997cheating the presence of semen. He also examined the sample microscopically. This examination did not reveal the presence of any spermatozoa, but did show a moderate amount of microorganisms. Gadd froze the sample on April 25 and it remained frozen until October 24, 1979.
In May 1979 the deputy district attorney assigned to try the case requested PGM typing on the swab. In June, however, she asked the lab to “hold up” further work. The PGM typing was not done in May, possibly because Gadd, in keeping with the custom and practice of the lab not to run PGM and ABO testing until blood and saliva samples were obtained from the victim and the suspect, was still waiting for a blood sample from the suspect. A blood sample had 'been obtained from the victim on April 20, 1979, but Gadd never received a saliva sample from her or a blood or saliva sample from the defendant. After defendant was arrested, Sergeant Bush, the investigating officer, had attempted to obtain a blood sample from him, but Mr. Hardeman, defendant’s then attorney, refused to permit Bush to contact defendant for the purpose of securing a blood sample. No court order was sought to compel production.
Mr. Hardeman was aware of the availability of the semen sample by the end of April 1979, but made no request for the specimen. Neither the semen sample nor the vaginal swab were introduced at defendant’s first trial in July 1979.
In October 1979, PGM typing was again requested. On October 24, 1979, Gadd removed the swab from the freezer and cut out a portion of it to use in a PGM test. The test was attempted on October 24 and 25, with inconclusive results. The test was repeated on a different portion of the swab on October 25 and 26, again without obtaining any useful results.
Sometime in October, after the inconclusive PGM tests, Sergeant Bush telephoned Gadd and indicated that he would be over to pick up the evidence. Gadd placed the vaginal swab aside in preparation for Bush’s picking it up, but Bush neglected to do so. The swab was left unfrozen from October 1979 until January 22, 1980.
In September 1979, Allen Ruby replaced Hardeman as defendant’s counsel. During September and October 1979, Ruby attempted to discover the status of the vaginal swab and whether any tests had been performed. Sometime in October Ruby learned that the laboratory results were inconclusive. In November or December he learned that the specimen had been unfrozen for a significant period of time, to an extent that any addi*998tional testing would be unlikely to yield useful results. Ruby filed the motion to dismiss in January 1980.
On January 22, 1980, subsequent to the hearing on the motion to dismiss, Gadd attempted an ABO grouping test. No useful result was obtained.
Custom and Practice of the Relevant Law Enforcement Agencies
The Sheriff’s Office
On receipt of a rape complaint, a deputy sheriff takes the complainant to the Valley Medical Center for a vaginal examination. Within one hour of receipt of the swab from the examining physician, the deputy places the evidence in the refrigeration unit of the medical section in the county jail. Refrigeration is maintained until the swab is delivered to the county criminalistics laboratory. After the sample has been analyzed by the laboratory, it is returned to the sheriffs department and booked into the unrefrigerated evidence locker at the sheriffs office.
The Criminalistics Laboratory
At the time relevant to this case, on receipt of the vaginal swab the criminalist would air-dry the swab to reduce the bacterial activity that can degrade the sample. In April 1979 air-drying was one of several acceptable means, although not the optimal means, of preserving the sample. The length of time necessary for air-drying varies with the viscosity of the sample, but two days is a normal length of time.
Usually the lab performs an initial analysis immediately after the sample has air-dried, and then places the sample in a box for retrieval by the law enforcement agency. However, in cases where there is a possibility that additional analysis will be required, the sample is stored in the laboratory’s freezer.
PGM and ABO grouping tests are not done routinely, but only if the laboratory receives some indication that the tests will be needed. It is the custom and practice of the laboratory not to perform the tests until it receives a blood and saliva sample from both the victim and the suspect. The reason for waiting for the blood and saliva samples is to determine if in fact it will be possible to differentiate between the PGM type or ABO group of the victim and suspect, so as to avoid wasting a valuable sample on a system that cannot be differentiated.
*999In 1979 it was not the practice of the Santa Clara County Criminalistics Laboratory to do ABO grouping on vaginal specimens because there were indications that the mixing of semen with vaginal secretions would preclude useful results. Although more recent research indicates that interpretable results can be obtained in a number of cases, the achievement of results is dependent upon whether the parties involved are secreters and, if so, whether they secrete enough of the antigen material to permit its detection.
Factors Affecting the Inability to Obtain Useful Results in the PGM Typing and ABO Grouping in this Case
A PGM type is a particular type of enzyme found in semen, blood and vaginal secretions. There are basically three common PGM types.
An ABO grouping is a person’s blood type. If a person is a secreter of the ABO antigens, the person’s ABO grouping can be determined from vaginal secretions or semen.
The PGM Type
Criminalist Gadd could not say with certainty that any particular factor was primarily responsible for the laboratory’s inability to obtain a useful result in the PGM testing conducted in October 1979. However, Gadd testified that the laboratory’s inability to obtain a useful PGM result could have been affected by any one or a combination of the following factors:
(1) The lapse of time between the act of intercourse and collection of the sample. The greater the delay, the less likely the sample is to yield a positive reading, and a delay of between 12 and 18 hours could affect the sample.4
(2) The presence of the moderate amount of microorganisms observed when the sample was examined microscopically on April 25, 1979. Gadd recognized then that the amount of microorganisms could create a problem for future processing, and stated that the amount of microorganisms could possibly have prevented obtaining interpretable results even if the swab had been taken and tested immediately. The microorganism activity that may have destroyed the integrity of the semen sample might have been increased by (a) the length of time between intercourse and collection of the sample, *1000(b) the dampness of the swab for a period of time, or (c) the six-month delay in performing the PGM testing.
(3) The amount of seminal material present on the swab.
(4) Whether one or both of the parties involved have sufficient amount of the PGM enzyme to permit typing.
(5) The delay in processing the sample. However, Gadd testified that for reasons not yet understood, in some instances vaginal specimens taken and tested immediately after intercourse will fail to yield decipherable results.
The ABO Group
Gadd similarly was unable to say why the ABO grouping attempted in January 1980 was unsuccessful. However, he testified that there were four possible contributing factors:
(1) The same deteriorative factors that prevented obtaining a useful PGM typing, including delay in collection of the sample, the microorganism activity, the amount of seminal material, and the delay in testing. However, the ABO antigen is more stable than the PGM enzyme and there is not necessarily a correlation between the success of the two types of tests.
(2) Whether the parties involved are secreters of the ABO antigens.
(3) The lack of refrigeration of the sample from October 1979 to January 1980. Gadd stated that this lack of refrigeration would degrade the sample for ABO grouping to some extent, but not to the same extent as it would for PGM typing. (Cf. People v. Nation (1980) 26 Cal.3d 169, 177 [161 Cal.Rptr. 299, 604 P.2d 1051] [where an unrefrigerated semen sample yielded a blood type when analyzed after the trial].)
(4) Additionally, an interpretable ABO grouping, like a PGM typing, is sometimes impossible to get even under optimal conditions.
Discussion
Was material evidence lost or destroyed as a result of failure by the prosecution to take reasonable precautions to preserve the evidence?
People v. Hitch (1974) 12 Cal.3d 641, 649, 652 [117 Cal.Rptr. 9, 527 P.2d 361], established that it is a requirement of due process of law *1001that the prosecution preserve evidence that comes into its possession if there is a reasonable possibility that the evidence would be material on the issue of the defendant’s guilt or innocence. The court held that when material evidence cannot be disclosed because of its intentional but nonmalicious destruction by the investigative officials, sanctions must be imposed for such nonpreservatioh and nondisclosure unless the prosecution can show that the governmental agencies involved have established, enforced and attempted in good faith to adhere to appropriate procedures designed to preserve the evidence. (At pp. 652-653.)
Was the Evidence Material?
Usually when a Hitch motion for sanctions is made, the defendant must first establish that the evidence is material. (Id., 12 Cal.3d at pp. 647-649; see, e.g., People v. Matthews (1980) 112 Cal.App.3d 11, 22-23 [169 Cal.Rptr. 263]; People v. Harris (1976) 62 Cal.App.3d 859, 864-865 [133 Cal.Rptr. 352].) When the evidence is no longer in existence, the defendant need only show that there is a reasonable possibility that the evidence, if preserved, would have constituted favorable evidence on the issue of guilt or innocence. (People v. Hitch, supra, 12 Cal.3d at p. 649.)
With respect to semen samples obtained from the vagina of an alleged rape victim, People v. Nation, supra, 26 Cal.3d 169, 175, 177, has established as a matter of law that the sample is material evidence requiring preservation. In the case at hand the moving party through the uncontradicted testimony of a deputy sheriff and a criminalist showed that on April 19, 1979, a vaginal smear of the complaining witness was taken at Valley Medical Center under the direction of the Santa Clara County Sheriff’s Department and that when tested on April 25, 1979, the swab showed the presence of seminal material. Once that factual showing was made the trial court was required to and did treat the sample as material evidence.
The People argue that the swab was material evidence only in the abstract. Observing that the two attempts to perform PGM typing in October yielded no useful results, they rely upon People v. Matthews, supra, 112 Cal.App.3d 11, to contend that where a swab discloses nothing of value, it cannot be found to be material in fact.
This contention sidesteps the issue. Assuming, arguendo, that when tested in October the swab’s failure to yield useful PGM results rendered it at that point immaterial in fact (although the possibility remains that a test for ABO group may have been successful), the question remains whether *1002its uselessness is attributable to the government’s failure properly to preserve the evidence. As the court stated in Nation, when evidence is lost to the defense because of its destruction by the authorities, it is “deemed material ... if there is a reasonable possibility that it would be favorable to the defendant on the issue of guilt or innocence.” (Id., 26 Cal.3d at p. 176.) Although the People assert that the swab’s uselessness was attributable to microorganisms present from the beginning, neither the record nor the findings support this unqualified statement.
Matthews, supra, 112 Cal.App.3d 11, is thus inapposite. In that case the nonpreserved ampoule deteriorated to an unrecognizable state quickly after use so that even had the inspector preserved it he could have produced “only a deteriorated material in an unrecognizable state. ” (Id., 112 Cal.App.3d at pp. 22-23.)
Moreover, in United States v. Bryant (D.C. Cir. 1971) 448 F.2d 1182 (cited on another point by the Supreme Court in People v. Hitch, supra, 12 Cal.3d at p. 654), the court held that the duty to preserve evidence applies to all discoverable evidence, and the tapes at issue in that case were no less discoverable because they were unintelligible; “It is the defendant’s right to discover such evidence and decide for himself . . . .” (United States v. Bryant, supra, 448 F.2d at p. 1184, fn. 1.)
Is the Loss of the Evidence Attributable to the Prosecution ?
Under Hitch and its progeny, the mere loss or unavailability of material evidence is insufficient of itself to justify imposition of sanctions; there must be a preliminary finding that the evidence was lost because of its destruction or inadequate preservation by the investigative officials. (See People v. Hitch, supra, 12 Cal.3d at p. 652; People v. Nation, supra, 26 Cal.3d at p. 176; see also People v. Maese (1980) 105 Cal.App.3d 710, 720 [164 Cal.Rptr. 485]; cf. Bellizzi v. Superior Court (1974) 12 Cal.3d 33, 36-37 [115 Cal.Rptr. 52, 524 P.2d 148], cert, den., 420 U.S. 1003 [43 L.Ed.2d 761, 95 S.Ct. 1445] [witness’ unavailability not attributable to prosecution].)
In this case the evidence was unavailable or “lost” because at the time of the PGM test in October and the ABO test in January the sample had deteriorated to the point where it could no longer yield useful results. It is unclear to what extent this inability is attributable to the government, particularly in light of the interval between the assault and the collection of the sample which, alone could have been sufficient to render the sample useless.
*1003 Did the prosecution meet its burden of establishing that it undertook reasonable efforts to preserve the material evidence?
Standard of Review
In granting defendant’s motion to dismiss, the trial court necessarily concluded that the prosecution had not met its burden of establishing that it had undertaken reasonable efforts to preserve the sample. (Cf. People v. Goss (1980) 109 Cal.App.3d 443, 454 [167 Cal.Rptr. 224].)
The question of whether the People have met their Hitch burden is a mixed question of law and fact; the substantial evidence test is applicable only to the trial court’s factual determinations.5 (Cf. People v. Leyba (1981) 29 Cal.3d 591, 596-598 [174 Cal.Rptr. 867, 629 P.2d 961] [lawfulness of detention]; People v. Jimenez (1978) 21 Cal.3d 595, 609 [147 Cal.Rptr. 172, 580, P.2d 672] [voluntariness of confession]; People v. Lawler (1973) 9 Cal.3d 156, 160 [107 Cal.Rptr. 13, 507 P.2d 621] [lawfulness of search and seizure].)
By analogy to the two-step process by which a trial court rules on a motion to suppress evidence under section 1538.5 (People v. Leyba, supra, 29 Cal.3d at p. 596), it can be said that the trial court’s determination whether the prosecution has met its Hitch burden is a two-step process: first, the court must “find the facts” i.e., whether the governmental agencies involved have established, enforced, and attempted in good faith to adhere to systematic procedures for preservation of evidence; second, the trial court must decide whether, on the facts found, the government’s procedures were legally sufficient to fulfill its duty under Hitch and Nation. This latter issue is a question of law with respect to which the appellate court has the “ ‘ultimate responsibility ... to measure the facts, as found by the trier, against the [appropriate legal standard].’ ” (Cf. People v. Leyba, supra, at pp. 596-597.) In the present case, the facts concerning the governmental procedures in effect in 1979 and the actions of the investigating authorities are largely undisputed.
*1004 Legal Sufficiency of the Procedures
In Hitch the court imposed on governmental agencies the duty to establish, enforce and attempt in good faith to adhere to “rigorous and systematic procedures” designed to preserve the evidence. (Id., 12 Cal.3d at pp. 652-653.) In Nation the court restated the Hitch rule as requiring law enforcement agencies to take “reasonable measures” to ensure the adequate preservation of material evidence. (People v. Nation, supra, 26 Cal.3d at p. 175.) Post -Nation cases have continued to refer to the “rigorous and systematic procedures” standard. (See, e.g., People v. Zamora (1980) 28 Cal.3d 88, 98 [167 Cal.Rptr. 573, 615 P.2d 1361]; People v. Bally (1981) 125 Cal.App.3d 584, 587 [178 Cal.Rptr. 96]; People v. Goss, supra, 109 Cal.App.3d at p. 455; People v. Jackson (1980) 102 Cal.App.3d 620, 623 [162 Cal.Rptr. 574].) Assuming, therefore, that the original formulation of the rule in Hitch remains the correct statement of the criterion, the question presented is whether the procedures established and followed in this case satisfy the rule.
Because the prosecution delayed in analyzing the evidence and left it unrefrigerated after the attempted PGM typing in October proved fruitless, resolution of this issue requires a determination of the scope of the duty imposed by Hitch and Nation, i.e., must the government, in the absence of a defense request and on its own initiative, undertake to analyze evidence which the defense knows it has, when there is reason to believe that the passage of time may by itself render the evidence lost or unavailable? Stated altemaltively, after the defense has been notified of the existence of the evidence, is a failure to test a failure to preserve?
The People argue that up until October 1979, when the sample was left unrefrigerated, the procedures adopted and followed by the investigating authorities (collection, refrigeration, air-drying, and freezing of the sample) were in keeping with then-acceptable scientific procedures; that there was no requirement that the prosecution criminalist, without any defense request, undertake PGM testing, or that the prosecution obtain a court order requiring the defendant to provide a blood sample to “enable them to perform such procedures”; and that once the state criminalist determined that the sample did not yield useful PGM results, absent a defense request there was no duty further to preserve it.
Defendant argues that the failure to perform ABO or PGM testing in April was unreasonable because the prosecution had actual knowledge that the semen sample might degrade over the passage of time; that the prosecution’s obligation to preserve the evidence was unconditional and *1005not dependent on the successful pursuit of a blood sample from defendant; and that leaving the sample unfrozen from October 1979 to January 1980 resulted from the lack of a system and thus was inexcusable, resulting in the loss of material evidence.
Underlying Hitch is defendant’s due process right to disclosure of potentially favorable material evidence. Hitch established that before a request for discovery has been made, the duty of disclosure is operative as a duty of preservation. (Id., 12 Cal.3d at p. 650.) Nation instructs that mere retention of evidence is insufficient if special measures are required to preserve its evidentiary value. (Id., 26 Cal.3d at p. 177.)
The underlying assumption of both cases is that once the defense is made aware of the existence of the evidence, it will request its production. Thus, in Hitch the court observed that if the duty of disclosure did not attach as a duty of preservation as soon as the government has gathered and taken possession of the evidence, “disclosure might be avoided by destroying vital evidence before . . . defendants hear of its existence.” (Id., 12 Cal.3d at p. 650.) Similarly, in Nation the court stated: “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. ” (Id., 26 Cal.3d at p. 175, italics added.)
Neither case appears to contemplate that the prosecution itself, in the absence of a defense request, is required to analyze the evidence. Thus, in Nation the court stated: “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.” (Id., 26 Cal.3d at p. 177.) Similarly, in Hitch the court analogized the prosecution’s duty to preserve material demonstrative evidence to its duty to preserve the testimony of a material informant. In the latter case, the duty to preserve is defined as a duty to “ ‘undertake reasonable efforts to obtain information by which the defense may locate such an informer.’ ” (Id., 12 Cal.3d at p. 650, italics added.) Significantly, the prosecution is not itself required to subpoena the witness; to the contrary, “[generally speaking the People may select and choose which witnesses they wish to use to prove their case against a defendant. ” (People v. Mejia (1976) 57 Cal.App.3d 574, 580 [129 Cal.Rptr. 192].)
*1006The rule that the prosecution is not required to produce a material informant witness is consistent with the general rule, cited by the People, that “the law does not impose upon law enforcement agencies the requirement that they take the initiative, or even any affirmative action, in procuring the evidence deemed necessary to the defense of an accused.” (In re Koehne (1960) 54 Cal.2d 757, 759 [8 Cal.Rptr. 435, 356 P.2d 179]; accord In re Martin (1962) 58 Cal.2d 509, 512 [24 Cal.Rptr. 833, 374 P.2d 801]; People v. Jenkins (1974) 40 Cal.App.3d 1054, 1056-1057 [115 Cal.Rptr. 622]; People v. Watson (1977) 75 Cal.App.3d 384, 399 [142 Cal.Rptr. 134]; People v. Hogan (1982) 31 Cal.3d 815, 851 [183 Cal.Rptr. 817, 647 P.2d 93].) Due process is offended only when the authorities, by their actions or regulations, impose any material obstacle in the path of the accused or frustrate his reasonable efforts designed to produce probative evidence. (In re Martin, supra, 58 Cal.2d at p. 512; cf. Bellizzi v. Superior Court, supra, 12 Cal.3d at p. 36 [dismissal required only when prosecution deprives accused of the opportunity to secure the presence of a material witness].)
We understand the Hitch-Nation duty to preserve material evidence as not embracing a correlative duty to test the evidence in the absence of a defense request; rather, the duty is to establish, enforce and attempt in good faith to adhere to reasonable procedures designed to preserve the evidence so that it will be available for testing by the defendant for a reasonable period of time. It follows that where, as here, after the prosecution has made a timely notification to the defense of the existence of the evidence and the defense fails in a timely fashion to avail itself of the evidence, subsequent deterioration of the evidence attributable to the passage of time cannot be said, as a matter of law, to be attributable to the prosecution. (Cf. People v. Nation, supra, 26 Cal.3d at pp. 177-178 [despite state’s initial failure to preserve the evidence, court refused to say that its deterioration was caused by state’s inaction when defense had subsequently also failed to preserve or analyze it]; People v. Watson, supra, 75 Cal.App.3d at p. 390, 400 [failure to perform tests requested by defense six weeks after the offense was due to nature of the substance, i.e., age of the blood stains, and not to any conduct on the part of the prosecution].)
Defendant, however, argues that in this case failure to test the sample immediately after its acquisition was unreasonable because the criminalist was then aware that the presence of microorganisms might degrade the sample over the passage of time. Because there were no established rules requiring immediate testing of samples found to contain a moderate amount of microorganisms, the narrow question is whether the failure of the authorities to adopt rules requiring immediate testing was a failure to comply with the Hitch-Nation standard of preservation.
*1007We find nothing in Hitch or Nation requiring such immediate prosecution initiative. Nor does it appear appropriate to impose such a rule on the basis of the record before us which discloses that defendant’s trial counsel was made aware of the existence of the evidence contemporaneously with its gathering.
The order dismissing the information is reversed.
Caldecott, P. J., concurred.
Unless otherwise indicated, all further statutory references are to the Penal Code.
On June 24, 1981, this court appointed Judge Flaherty as referee to take additional evidence and make recommended findings. (Cal. Rules of Court, rule 23(b).) The referee’s findings were filed on February 2, 1982. At the direction of the court, the parties thereafter submitted supplemental briefs.
A fair reading of referee’s finding No. 1 indicates that the period was 26.5 hours.
The literature indicates that the likelihood of obtaining an accurate PGM typing from a swab is directly related to the lapse, of time between intercourse and collection of the sample, and that PGM activity decreases rapidly when swabs are taken more than two hours after intercourse. (Blake & Sensabaugh, Genetic Markers in Human Semen: A Review (1976) 21 J.Forensic Sci. 784, 794-795 (cited in People v. Wilson (1982) 128 Cal.App.3d 132, 136-137, fn. 3 [179 Cal.Rptr. 132]).
Insofar as the trial court’s implied findings are superceded by the referee’s express findings, the standard of review is that although the referee’s findings are entitled to great weight, they are not binding on the appellate court, which is required to make an independent examination of the record. (In re Anderson (1971) 6 Cal.3d 288, 297 [98 Cal.Rptr. 825, 491 P.2d 409]; People v. Chapman (1971) 5 Cal.3d 218, 225 [95 Cal.Rptr. 533, 485 P.2d 1149]; People v. Tucker (1964) 61 Cal.2d 828, 831 [40 Cal.Rptr. 609, 395 P.2d 449].)