MALLANO, J.

Jeffrey Mitchell and Robert Rogers appeal from judgments entered following a joint trial in which they were convicted of first degree murder and found to have personally used firearms. It was further found that the crime was committed for the benefit of a street gang. In the published portion of this opinion, we conclude that the trial court erred in permitting evidence that Mitchell had been identified in a canine scent identification lineup without first holding a hearing under People v. Kelly (1976) 17 Cal.3d 24. Nevertheless, we hold that the error was harmless. In the unpublished portion, we reject arguments that challenge Rogers's photographic identification and the propriety of certain evidence. Accordingly, we affirm.

FACTUAL BACKGROUND 1. The Crime and the Investigation

Defendants are members of the Underground Crips (UG Crips) gang. The intersection of 106th Street and Budlong Avenue in Los Angeles County is in the territory of the adjacent rival, Hoovers gang. Around 10 a.m. on August 20, 1999, 19-year-old Jose Lazaro was in his mother's house at the intersection of 106th and Budlong when he heard loud noises that sounded like firecrackers. After a pause, he heard two more similar sounds, and then about five more bangs which sounded more like gunshots. Lazaro looked through the black metal front screen door to investigate the sounds. For three or four seconds he was able to see two men, whom he later identified as defendants, on the sidewalk directly in front of the house. They were walking very quickly. As they walked, men later identified as Mitchell and Rogers bumped into each other. Mitchell was in front holding a gun. Lazaro testified that he was able to get a view of the right profile of both men and also a "partial look at the front of [Mitchell's] face when he turned around."

When Mitchell and Rogers proceeded a bit farther up the street, Lazaro went outside and saw the body of Peter Drake lying faceup in the intersection. Lazaro recognized Drake as a member of the Hoovers gang, which frequently congregated in a nearby parking lot. Lazaro went back inside and called 911 to report the shooting.

Paramedics and police officers soon arrived. Drake was declared dead at the scene. Lazaro did not go back outside to talk to the officers because he did not want to be recognized by neighbors. Instead, he again called 911 and gave descriptions of both assailants. In this call, Lazaro described the man with the gun (Mitchell) as African-American, 16 to 23 years old, about 5 feet 8 inches tall, with a "fade" haircut, and wearing a white shirt, black shorts, and black shoes with a white Nike insignia. Rogers was described as a bit taller and heavier, with slicked-back hair, and wearing white shoes with a black Nike insignia. Later that day, Lazaro worked with a police artist, who made a drawing of Mitchell that depicted what Lazaro described as the assailant's prominent cheekbones and Adam's apple. Lazaro had been in the Los Angeles County Sheriff's Department Explorer Program for two and one-half years and had been trained to notice such details. He testified that he has glasses with a correction for distance that he uses sometimes for driving at night but that he does not use them at any other time. He was not wearing the glasses when he saw the two assailants walk by.

Officers at the scene of the shooting located five bullet casings from a .25-caliber semiautomatic weapon near Drake's body. An autopsy revealed that Drake had sustained multiple gunshot wounds, with fatal or potentially fatal bullets lodged in his brain, neck, lung, and chest. Eight bullets were recovered, four of which were .25-caliber and had been fired from a semiautomatic gun. The other four bullets were .22-caliber and most likely had been fired from a revolver inasmuch as no casings of that caliber bullet were found at the scene.

On August 23, 1999, officers arrested UG Crips member Marcus Hartfield for burglary. Hartfield told the officers that he had some information about a murder. He stated that on August 20 he was at the home of UG Crips member Roderick Hyatt, who lived very close to 106th and Budlong but still in UG Crips territory. There, Hartfield saw Rogers and Mitchell speaking with senior UG Crips member Lynn Woods. Rogers told Woods that he (Rogers) had just shot a "Snoover" in the head. Mitchell added that he and Rogers had "`caught a Hoover slipping,'" and both defendants bragged that they had "`laid some Hoover down this morning.'" (A gang expert testified that "Snoover" is a derogatory term for Hoover and having "caught the victim slipping" means that a rival gang member had been caught off guard in his own territory.) Hartfield was released from custody a few hours after he gave this information.

Following the interview with Hartfield, officers put together three six-pack photographic arrays, which were shown to Lazaro on August 24. The target photographs in the arrays were of Rogers, Mitchell, and UG Crips member Ray Owens, respectively. The photograph of Rogers displayed only his head; it had been enlarged so that the plaid shirt he was wearing would not be seen. The photograph of Mitchell, who had turned 16 the week before the murder, had been taken in 1997. Lazaro did not make any identification from the array that had the photograph of Rogers. Viewing the next array, Lazaro stated that Mitchell "looks like" the person with the gun. He also told officers that another person in the array had the hair and cheekbones that resembled the person with the gun although he was not one of the perpetrators. Looking at the third array, Lazaro stated that the photograph of Owens also resembled the person with the gun but was not that person.

On September 10, 1999, Mitchell was arrested on an unrelated charge and a search warrant was executed at his home. Among other items, officers found four Winchester .25-caliber semiautomatic bullets (two of the .25-caliber casings found at the murder scene were Winchester brand) and a sheet of paper with writing that appeared to be lyrics of a rap song. The lyrics included the phrases: "`Caught off brand slipping'" and "`Put two in his brain.'" (The gang expert explained that "off brand" referred to a rival gang member.)

On September 11, Mitchell gave a statement to the police. Although he acknowledged his membership in the UG Crips gang and his gang's rivalry with the Hoovers, he disclaimed any knowledge of the murder. He also denied having been at the intersection of 106th and Budlong, a place he regarded as enemy territory. When asked where he had been at the time, Mitchell said he may have been in school or at the graduation of a fellow gang member. Mitchell further stated that the rap lyrics had been written shortly after his release from camp community placement on July 6, 1999, and had nothing to do with the Drake murder the following August 20.

Meanwhile, officers assembled front and profile photographs of 12 individuals, including Mitchell and Rogers (Rogers was photographed upon being detained in a gang sweep), all wearing identical white T-shirts and standing against the same background. Ray Owens was not included in the group of 12 because officers had been unable to detain him. Other than Mitchell and Rogers, none of the subjects included in the three six-packs previously shown to Lazaro were included in the new set of photographs.

The photographs were shown to Lazaro on September 12. Lazaro identified Mitchell as the person with the gun. He tentatively identified Rogers, saying that he would be definite if the person were a bit heavier and had a different hairstyle.

Rogers was arrested on September 12, 1999, also on unrelated charges. He denied any involvement in or knowledge of the murder. He later stated that on the morning of the murder a Hoovers gang member shot at him from a car and that he had told his fellow UG Crips gang members about the shooting. Rogers also stated that he knew there had been "walk-up shooting" on August 20 and that the "correct Hoover" had been killed. Rogers further admitted that he had gone to Roderick Hyatt's house on the day of the murder. When an officer said that it was unlikely that anyone could go into rival gang territory on foot, Rogers replied, "`That's exactly what they did. They walked over and they ran back.'" At one point during the interview, Rogers commented that "`[p]eople get murdered around here all the time'" and added, "`Why are you all doing this to me? Why me? Why this murder? Why do you have to solve this one?'"

On December 1, 1999, Lazaro attended live lineups that included Mitchell and Rogers. He identified both. A detective who attended the lineups testified that afterward Lazaro "came up to me, said `You should have done this the first time.' He said it was easy to make the identification[s]. That he was absolutely sure these were the suspects" and that Mitchell was the one with the gun. Lazaro reiterated his identifications at trial.

Finally, the prosecution placed into evidence Mitchell's "juvenile hall entrance record" that had been created upon his arrest on September 10, 1999. The record indicated the existence of one gang tattoo on each arm. The prosecution also introduced a photograph of Mitchell taken on February 5, 2001, during the course of trial, which showed an additional tattoo, of the initials "HK," on one of Mitchell's arms. The gang expert testified that the letters stood for Hoover killer.

2. Scent Identification Lineup Evidence

The prosecution also introduced evidence that Mitchell's scent had been identified in a lineup as being present on the expended shell casings found at the scene. The evidence was presented through the testimony of Joe D'Allura and Edward Hamm, civilian canine handlers who volunteer for, among other organizations, the Los Angeles County Sheriff's Department and the California Rescue Dog Association (CARDA). D'Allura handles Reilly, a nine-year-old Labrador retriever that performed the scent identification lineup in this case. Hamm handles bloodhounds. Hamm's dogs were not used in this case, but he assisted D'Allura and Reilly in the scent lineup in which Mitchell's scent was identified.

D'Allura is a trainer of "article dogs, cadaver dogs, and scent discrimination dogs." Scent discrimination as performed by Reilly is to be distinguished from trailing scents or tracking, which is typically done by bloodhounds. Hamm, D'Allura, and their dogs often work as partners in the field, with Hamm's bloodhound following the scent trail of a person and Reilly following behind, seeking items of evidence.

D'Allura has worked with Reilly since Reilly was a puppy and has been Reilly's only handler. Reilly was first trained in search and rescue operations to find people who are lost or hiding. After hundreds of hours of training and passing a test in which he found two people within an hour who were concealed in a 160-acre area, Reilly received CARDA certification in search and rescue. Reilly next was trained to conduct cadaver searches, in which he received CARDA certification upon passing a test requiring that within a 20-minute period he find a cadaver that was buried at least 18 inches below ground level. Next Reilly was trained to conduct article searches, in which he located hidden or buried articles that had been touched by human beings. Certification in that area was conducted by the sheriff's department. In handling Reilly during these training operations, D'Allura is never told where the person, cadaver, or article is hidden.

In 1997, Reilly began to work in the area of scent discrimination. This involves Reilly sniffing the scent collected from an object a person is known to have touched and determining whether a second object has been touched by the same individual. The scents Reilly sniffs have been collected by a small vacuum cleaner-like device known as a "scent transfer unit." The scent transfer unit, which was described by Hamm as "essentially a modified dust buster," was developed in 1994 by a bloodhound handler associated with the Niagara County Sheriff's Department in New York. It extracts scents from an object and transfers the scents to a sterile gauze pad, thus avoiding possible damage to the object from which the scent has been taken.

In training operations, the scent transfer unit is used to prepare five or six sterile pads. Two of the pads contain the scent of a target person and three or four pads contain the scent of others. When D'Allura and Reilly are out of the room, one of the pads that has the scent of the target person and the three or four pads that have the scent of others are put on the floor. D'Allura and Reilly then enter the room. Reilly initially sniffs the other pad that has the scent of the target person. After Reilly has sniffed this pad, D'Allura has Reilly walk by the pads lined up on the floor and sniff them. If Reilly perceives a match between the scent on the pad he was initially given to sniff and one of the pads lined up on the floor, he alerts D'Allura by stopping and barking. If Reilly selects the pad with the scent of the target person, Reilly is given a reward. In lineups involving actual suspects the protocol is the same, except that only one pad is known to contain the scent of the suspect, and Reilly's task is to determine if scent on that pad matches scent on a pad taken from an object the suspect is believed to have touched.

Reilly was trained in scent discrimination for several hundred hours and in 1999 was "certified in this area as well." To keep his certification, Reilly continues to train and perform scent lineups on a weekly basis. As of the date of D'Allura's testimony, Reilly had performed over 200 training lineups and over 100 lineups involving actual suspects. Since his certification, Reilly had not made a mistake in any training lineup, and there has been no indication that Reilly had made any mistakes in an actual lineup. Training lineups in which Reilly participated included one where a person handled a soda bottle and a gas can that were put in a car that was doused with gasoline, set on fire, and allowed to burn for 20 minutes. When the fire was extinguished, scents were extracted from the objects, which had then "melted into blobs." Reilly was able to identify a person who had touched the objects before the fire in a scent pad lineup. In another training exercise, Reilly matched the scent of a person with the scent of coins the person had carried in his pocket for a day. The lineup was conducted after the coins had been heated in an autoclave, which is a device used for sterilizing surgical instruments. In both training and actual lineups, Reilly has successfully matched scent collected by the scent transfer unit from expended shell casings to the person who put the bullets into the gun from which they were fired.

D'Allura, who has no background or education in science, explained it was his understanding that the scent detected by dogs comes from microscopic skin cells, known as scruff. According to D'Allura, human "[s]cent is like a fingerprint, basically. Everybody has their own scent about them." Scent will remain on an object for two to four months after it has been touched. (This is to be contrasted with tracking or trailing, which is typically done by a bloodhound. There the dog's main focus is to follow the scent of skin cells that are in the air or have fallen to the ground. This type of scent is more susceptible to dissipation from weather and the passage of time.)

In the case at bench, Hamm used the scent transfer unit to collect scents from the murder shell casings, which had been kept in an envelope, as well as from the victim's shirt. (The prosecution theorized that Mitchell had turned the victim over after he was already on the ground.) Scents were also collected from shirts Mitchell and Rogers had worn, and additional scents were collected from shirts worn by other UG Crips members. Hamm also made three control pads with scents extracted from the chairs of three detectives in the sheriff's homicide bureau who had not worked on the case. The protocol employed in each instance of scent collection was that Hamm, wearing surgical gloves, would insert a sterile gauze pad in the scent transfer unit, turn the unit on, place the unit for about one minute over the object from which scent was to be extracted, turn the unit off, and remove the gauze pad, placing it into a "Ziploc bag." Hamm then would clean all contact surfaces of the scent transfer unit with alcohol and run it briefly to dry the alcohol.

On September 24, 1999, D'Allura and Reilly participated in nine lineups that had been set up by Hamm. In the first lineup, Reilly was given a pad to sniff that had scent extracted from the shell casings and matched it in the lineup to the pad that contained scent collected from Mitchell's shirt. In another lineup, Reilly was given a pad to sniff that had scent collected from Drake's shirt and matched it in the lineup with Mitchell's pad. Scent taken from Mitchell was not involved in any of the remaining seven lineups. Reilly did not alert during these lineups, which included scent taken from Rogers, that any matches had been found.

3. Defense

Neither defendant testified. In Mitchell's defense, UG Crips member Lynn Woods denied having had a conversation with defendants in which they admitted the shooting. An eyewitness identification expert testified on factors affecting cross-racial identification (Lazaro is Hispanic, defendants are African-American). The expert was of the opinion that the identifications in this case were also affected by Lazaro's limited opportunity to see the perpetrators and that the lineup procedure in which Lazaro was repeatedly exposed to Mitchell's photographs was unfair.

Rogers presented an alibi defense. Marcus Ward and Laquandra West testified that Ward, who was Rogers's best friend, graduated from an alternative high school on the day of the murder. Between 9:00 and 9:30 that morning, Ward drove Rogers to West's house to see if she was coming to the graduation and then drove Rogers to Ward's house so that Rogers could take a shower and get dressed. Thereafter, the two left Ward's house, arriving sometime between 10:00 and 10:30 a.m. Rogers and West then walked to the home of some friends, after which they went to the high school, arriving between 10:30 and 11:00 a.m. From the school, they went to a Sizzler restaurant, where the graduation ceremony and party took place, arriving at around noon. The principal of the school testified that she first saw Ward and Rogers at the high school between 10:30 and 10:45 a.m.

In rebuttal, a detective testified that Ward told him that Rogers first arrived at Ward's house between 10:00 and 11:00 a.m. on the day of the murder. Ward's mother told another detective that Ward was at home throughout that morning, first leaving the house to go to the Sizzler. Ward's mother also said that she had received a telephone call from West that morning. In the call, West acted hysterically and stated that Rogers was trying to get her to cover up something.

DISCUSSION 1. Scent Identification Lineup Evidence

Before trial, Mitchell moved under Evidence Code section 402 to exclude evidence of the scent identification lineup on grounds that an adequate foundation could not be established for either scent identification lineup evidence in general or use of the scent transfer unit in particular. Hamm and D'Allura testified at a hearing on the motion. That testimony, which has been summarized above, presaged their testimony at trial. The trial court denied Mitchell's motion, finding that analysis under the Kelly-Frye rule did not apply to the challenged evidence and that a sufficient foundation had been established under existing case law. We conclude that this ruling was in error.

a. Relevant Legal Principles

"In People v. Kelly[, supra,] 17 Cal.3d 24 . . . ( Kelly), [the Supreme Court] held that evidence obtained through a new scientific technique may be admitted only after its reliability has been established under a three-pronged test. The first prong requires proof that the technique is generally accepted as reliable in the relevant scientific community. ( Id. at p. 30.) The second prong requires proof that the witness testifying about the technique and its application is a properly qualified expert on the subject. ( Ibid.) The third prong requires proof that the person performing the test in the particular case used correct scientific procedures. ( Ibid.) [The Supreme Court] further held that proof of a technique's general acceptance in the relevant scientific community would no longer be necessary once a published appellate decision had affirmed a trial court ruling admitting evidence obtained by that scientific technique. . . ." ( People v. Bolden, supra, 29 Cal.4th at pp. 544 545.)

" Kelly is applicable only to `new scientific techniques.' [Citations.]" ( People v. Leahy (1994) 8 Cal.4th 587, 605.) It "`only applies to that limited class of expert testimony which is based, in whole or part, on a technique, process, or theory which is new to science and, even more so, the law.' [Citation.]" ( Ibid.) As stated by the Leahy court in discussing People v. Stoll (1989) 49 Cal.3d 1136, "by reason of the potential breadth of the term `scientific' in the Kelly/Frye doctrine, the courts often refer `to its narrow "common sense" purpose, i.e., to protect the jury from techniques which . . . convey a "`misleading aura of certainty.'" [Citations.]' ( 49 Cal.3d at pp. 1155 1156.) According to Stoll, a technique may be deemed `scientific' for purposes of Kelly/Frye if `the unproven technique or procedure appears in both name and description to provide some definitive truth which the expert need only accurately recognize and relay to the jury.' ( Id. at p. 1156, italics added.)" ( People v. Leahy, supra, 8 Cal.4th at p. 606.)

As explained in People v. McDonald (1984) 37 Cal.3d 351: "When a witness gives his personal opinion on the stand — even if he qualifies as an expert — the jurors may temper their acceptance of his testimony with a healthy skepticism born of their knowledge that all human beings are fallible. But the opposite may be true when the evidence is produced by a machine: like many laypersons, jurors tend to ascribe an inordinately high degree of certainty to proof derived from an apparently `scientific' mechanism, instrument, or procedure. Yet the aura of infallibility that often surrounds such evidence may well conceal the fact that it remains experimental and tentative. [Citation.] For this reason, courts have invoked the Kelly-Frye rule primarily in cases involving novel devices or processes such as lie detectors, `truth serum,' Nalline testing, experimental systems of blood typing, `voiceprints,' identification by human bite marks, microscopic analysis of gunshot residue, and hypnosis [citation], and, most recently, proof of guilt by `rape trauma syndrome' [citation]. In some instances the evidence passed the Kelly-Frye test, in others it failed; but in all such cases `the rule serves its salutary purpose of preventing the jury from being misled by unproven and ultimately unsound scientific methods.' [Citation.]" ( People v. McDonald, supra, 37 Cal.3d at pp. 372 373, overruled on another ground in People v. Mendoza (2000) 23 Cal.4th 896, 914.)

Thus, Kelly analysis is limited to situations where it will "forestall the jury's uncritical acceptance of scientific evidence or technology that is so foreign to everyday experience as to be unusually difficult for laypersons to evaluate. [Citation.] In most other instances, the jurors are permitted to rely on their own common sense and good judgment in evaluating the weight of the evidence presented to them. [Citations.]" ( People v. Venegas (1998) 18 Cal.4th 47, 80.)

Finally, regardless of whether evidence is deemed "scientific," it will not be admitted unless it is relevant. In Daubert v. Merrell Dow Pharmaceuticals, Inc., supra, 509 U.S. at page 597, the United States Supreme Court determined that the Frye standard had been superseded by the Federal Rules of Evidence because "`[g]eneral acceptance' is not a necessary precondition to the admissibility of scientific evidence" under the Federal Rules, which "assign to the trial judge the task of ensuring that an expert's testimony both rests on a reliable foundation and is relevant to the task at hand. Pertinent evidence based on scientifically valid principles will satisfy those demands." In People v. Leahy, supra, 8 Cal.4th 587, the California Supreme Court concluded that Daubert did not provide cause for this state to abandon Kelly. Nonetheless, the Leahy court recognized that provisions of our Evidence Code "seem the functional equivalent" of the Federal Rules of Evidence relied on in Daubert. ( People v. Leahy, supra, 8 Cal.4th at p. 598.)

In California evidence is relevant only if it has "any tendency in reason to prove or disprove any disputed fact" (Evid. Code, § 210). And an expert's testimony must be based on matter "that is of a type that reasonably may be relied upon by an expert" ( id., § 801, subd. (b)). (See People v. Leahy, supra, 8 Cal.4th at pp. 597 598.)

i. California Case Law

There is no California case that involves a scent identification lineup. There are three that involve tracking or trailing by scent, and the trial court in this case applied those opinions by analogy to the scent identification lineup here. In the first, People v. Craig (1978) 86 Cal.App.3d 905, three men in a white Nova robbed a gas station. Station employees pursued the men to an apartment complex, where the men stopped and ran inside. When police officers arrived at the complex, they saw three men who matched the descriptions of the suspects and ultimately detained them. ( Id. at pp. 909 910.) The robbery victim was brought to the scene, where he identified the defendant. When the Nova was searched, incriminating evidence from a separate robbery which had occurred earlier that evening was found inside. ( Id. at pp. 910 911.) An officer and his "trained police dog" ( id. at p. 910) "were ordered to track from the interior of the vehicle. After being allowed to smell inside the Nova, the dog followed the path of the suspects from that point to the point where the detention occurred." ( Id. at p. 911.)

The Craig court held this evidence of canine tracking was admissible and not subject to Kelly. It observed that "[i]n the area of new scientific techniques, especially dealing with electronic gadgetry, one piece of testing apparatus is essentially the same as another of similar design, make and purpose. [¶] When dealing with animate objects, however, we must assume each and every unit is an individual and is different from all others. . . . [¶] Therefore, rather than attempt to identify certain specific criteria as being indicative of the ability of dogs, in general, to trail a human, we choose to require each particular dog's ability and reliability to be shown on a case-by-case basis. We are not merely assuming a well-trained dog can trail a human; we say that this ability is a fact which, like other facts, may be proven by expert testimony. [¶] This testimony should come from a person sufficiently acquainted with the dog, his training, ability and past record of reliability. If the testimony comes from an expert in the area of training, trailing, and operational performance of such dogs, that expert is qualified to state an opinion as to the ability of that particular dog in question to trail a human. We feel People v. Kelly, supra, and People v. King (1968) 266 Cal.App.2d 437, 458 . . ., are distinguishable in that both cases deal with problems of general acceptance in the scientific community of inanimate scientific techniques, rather than specific recognition of one animal's ability to utilize a subjective, innate capability." ( People v. Craig, supra, 86 Cal.App.3d at pp. 915 916, fn. omitted.) The Craig court found the dog handler in that case to be an expert and that the dog in question had been sufficiently trained to render tracking evidence admissible. ( Id. at pp. 916 917.)

In People v. Malgren (1983) 139 Cal.App.3d 234, 237, victims returned to their home one evening, noticed that items had been moved, heard a loud noise in a bedroom, and saw someone run down the hall and out the rear door into the backyard. An officer and his tracking dog responded to the victims' call, arriving at the house less than a half-hour later. From inside the house, the dog was commanded to "track." The dog ran down the hallway, into the bedroom, and out the same door as the suspect, continuing across the backyard into an adjacent game reserve. The dog "tracked for approximately 35 minutes and over about seven-tenths of a mile, and then ran into some high bushes and began to growl and bite. [The defendant] was found in the bushes, out of breath and perspiring." ( Ibid.)

Citing an American Law Reports annotation captioned Evidence of Trailing by Dogs in Criminal Cases, which noted that such evidence was admissible in a majority of jurisdictions (Annot. (1968) 18 A.L.R.3d 1221, superseded by Annot. (2000) 81 A.L.R.5th 563), the Malgren court observed that the holding in People v. Craig, supra, 86 Cal.App.3d 905, was consistent with the majority view on admissibility. ( People v. Malgren, supra, 139 Cal.App.3d at pp. 237 238.) The Malgren court continued: "While we agree that in each case the proponent of dog tracking evidence must establish the dog's ability and reliability, we believe a proper foundation must also include evidence that the circumstances of the tracking itself make it probable that the person tracked was the guilty party [citation]. We conclude that the following must be shown before dog trailing evidence is admissible: (1) the dog's handler was qualified by training and experience to use the dog; (2) the dog was adequately trained in tracking humans; (3) the dog has been found to be reliable in tracking humans; (4) the dog was placed on the track where circumstances indicated the guilty party to have been; and (5) the trail had not become stale or contaminated." ( Id. at p. 238.)

Finally, in People v. Gonzales (1990) 218 Cal.App.3d 403, as in Malgren, a qualified handler and tracking dog were put on the trail of a burglary suspect within a half-hour after he had fled. The suspect was found hiding in an unplowed vineyard less than a mile from the home he had burglarized. ( People v. Gonzales, supra, 218 Cal.App.3d at pp. 405 406.) The defendant did not raise an issue regarding the dog tracking evidence. Rather, he argued that an instruction given to the jury which paralleled the five prerequisites for such evidence set forth in Malgren was inadequate. ( People v. Gonzales, supra, 218 Cal.App.3d at pp. 407 408.) The Gonzales court held that dog tracking evidence must be corroborated but did "not find the other corroborative evidence must, as a matter of law, be evidence which, standing alone, independently links the accused to the crime; the corroborative evidence need only support the accuracy of the tracking itself." ( Id. at p. 414.)

ii. Case Law of Other Jurisdictions

In State v. Roscoe (1985) 145 Ariz. 212 [ 700 P.2d 1312], the defendant abducted the bicycle-riding victim and drove her in his car to a location where he sexually assaulted and murdered her. ( Id. at pp. 1315 1316.) A dog performed several scent identifications, including selecting the defendant's car from a group of five cars as matching the scent of the victim's clothing and selecting the victim's bicycle from a group of five bicycles as matching the defendant's scent. ( Id. at p. 1318.) The court stated: "We do not believe that the Frye test is applicable to evidence of dogtracking or scenting. The evidence here was not bottomed on any scientific theory. In fact, it appears that no one knows exactly how or why some dogs are able to track or scent, or the degree to which they are able to do so. No attempt was made to impress the jury with the infallibility of some general scientific technique or theory. Rather, this evidence was offered on the basis that it is common knowledge that some dogs, when properly trained and handled, can discriminate between human odors." ( Id. at pp. 1319 1320, fn. omitted.)

In Winston v. State (Tex.App. 2002) 78 S.W.3d 522, a sheriff's department bloodhound tracked the scent from an object that had been disturbed during a burglary to the defendant's front door. The defendant later gave police a scent sample on a gauze pad. (There is no mention of how that sample was taken.) The pad was placed with four others, and the original bloodhound and a second bloodhound were also given the scent from the burglarized premises. Both alerted to the pad with the defendant's scent. ( Id. at p. 524.)

Addressing the contention that the foregoing evidence should not have been admitted, the court applied the test of "whether the field of expertise is a legitimate one." ( Winston v. State, supra, 78 S.W.3d at p. 526.) The court found that it was, stating that "[t]he ability of certain breeds of dogs, especially bloodhounds, to distinguish humans by scent is well-documented. [Citations.] Further, these dogs' superior senses have long been used to aid mankind in a variety of contexts outside the courtroom, including `to track by scent escaped criminals or lost persons and articles.' [Citations.]" ( Ibid.) After noting that 37 states and the District of Columbia "admit scent-tracking evidence to prove the identity of the accused, provided a proper foundation is laid," the Winston court stated: "[W]e believe there is little distinction between a scent lineup and a situation where a dog is required to track an individual's scent over an area traversed by multiple persons." ( Id. at p. 527.) The Winston court concluded that the foundation requirement for scent identification lineups would be that "the dog was given the scent within the period of its efficiency." ( Id. at p. 528.)

b. Analysis i. Scent Transfer Unit

We first address Mitchell's argument that regardless of whether scent identification lineups are themselves subject to Kelly, the scent transfer unit used in the preparation of the gauze pads employed in the scent lineup is a novel device within the meaning of Kelly. (See People v. McDonald, supra, 37 Cal.3d at p. 372.) The contention has merit.

Use of the scent transfer unit has not been discussed in a published opinion in California or any other jurisdiction. Citing People v. Bury (1996) 41 Cal.App.4th 1194 and People v. Nolan (2002) 95 Cal.App.4th 1210, the Attorney General argues that the lack of precedent is of no moment because Kelly "applies to scientific principles, methodologies and techniques, not the specific devices used to implement them." As we shall see, while analysis under Kelly may not be required for new devices used to conduct an established type of test or an established device used to conduct a new type of test, a scent lineup involves neither.

In People v. Bury, supra, 41 Cal.App.4th 1194, a preliminary alcohol screening (PAS) of the defendant was conducted by having him blow into an "`Alco-Sensor III'" machine. ( Id. at pp. 1199 1200.) The defendant challenged his arrest for driving under the influence of alcohol on the ground, among others, that "the `fuel cell' device in the Alco-Sensor machine" was a "new scientific development," which required a Kelly hearing. ( Id. at p. 1201.) In rejecting this contention, the Bury court observed that "[a] PAS device is characterized by the purpose of its use rather than the nature of the device itself. [Citation.] . . . Many types of breath-testing devices exist; by 1989 there were 56 approved PAS devices. [Citation.] [The defendant's] theory of requiring each new breath-testing device or mechanism to be subject to a Kelly/Frye analysis would lead to unnecessary and unduly burdensome litigation." ( Id. at p. 1202.)

In People v. Nolan, supra, 95 Cal.App.4th 1210, the defendant submitted a urine sample as part of a condition of her drug court probation, the sample was tested with an ADX Abbott Bench Analyzer, the results were positive, and the defendant was found in violation of probation. ( Id. at pp. 1212 1213.) The Nolan court rejected the defendant's contention of error based on the trial court's refusal to conduct a Kelly hearing. The court concluded that "[t]he ADX Abbott Bench Analyzer is a machine which tests urine samples to determine drug usage. Urinalysis is not new. It is a medically accepted and widely used method of drug testing. . . . [¶] . . . [¶] There will always be new devices to implement established scientific methods. But a Kelly/ Frye hearing is not required for new devices; it applies to new methodologies. [Citation.]" ( Id. at pp. 1214 1215.)

Also pertinent to this issue is People v. Mendibles (1988) 199 Cal.App.3d 1277. There, a medical doctor examined the external genitalia of prepubescent victims of sexual abuse utilizing a colposcope with an attached camera, which was described as a binocular device that magnifies the area being examined and permits the results of the examination to be photographically preserved. ( Id. at pp. 1287, 1292.) The defendant's contention that the results of that examination were subject to Kelly/ Frye was rejected on the ground that it was "clear there was no novel device involved. The colposcope is an instrument in general use in the medical community which has value in detecting sexual abuse or rape. . . . [¶] . . . The sole `novelty' apparent here is the analysis of injury to a specific portion of the external genitalia of prepubescent females. . . ." ( Id. at p. 1295; accord, People v. Luna (1988) 204 Cal.App.3d 726, 733.)

Unlike a colposcope, the scent transfer unit is a device that does not have a history of use in the field of law enforcement or in any other arena. And unlike the Alco-Sensor machine and the ADX Abbott Bench Analyzer, which were themselves new devices, the scent transfer unit is not being put to a well-established use such as measuring blood-alcohol level through breath or the presence of drugs in urine.

The Attorney General argues that the scent transfer unit "is simply based on the obvious principle that scent travels in air. There is no novel scientific principle behind the use of suction to move air and the scent contained in it; it is the same principle at work in air filters in every home." The argument begs the question. The technique of establishing a person's identity in a canine scent lineup has not been subject to judicial scrutiny in this state. Our common knowledge that scent travels through air and that vacuum devices pick up particles does not make transferring scent with a "modified dust buster" any less novel.

And it is far from obvious that, assuming a vacuum device would transfer scent to a gauze pad under the protocol used in this case, that scent would not degrade or become contaminated. Indeed, Hamm testified that after scent had been collected, he would clean the contact surfaces of the scent transfer unit with alcohol and run the unit to let the alcohol dry. But, as we have noted, Reilly's training included being able to identify a person's scent on objects that had been burned beyond recognition or sterilized in an autoclave. Hamm did not address the possibility that scent from an object that had been transferred to a gauze pad through the unit would remain in the unit after the cleaning process and would then be transferred to subsequent gauze pads on successive uses. As reported in The Baltimore Sun in an article regarding California bloodhounds brought to the East Coast to investigate the anthrax incidents of 2001, the Law Enforcement Bloodhound Association and the National Police Bloodhound Association have not endorsed the scent transfer unit, saying "it offers little advantage over using a gauze pad alone and in fact might confound matters." (Shane, FBI's Use of Bloodhounds in Anthrax Probe Disputed, The Baltimore Sun (Oct. 29, 2002) p. 1A.)

The scent transfer unit is a novel device used in furtherance of a new technique. (See People v. Leahy, supra, 8 Cal.4th at p. 605; People v. McDonald, supra, 37 Cal.3d at p. 373.) It was error to deny Mitchell's request that the device be subject to a hearing pursuant to Kelly.

ii. Scent Identification Lineup

We find the approaches taken in jurisdictions which have uncritically accepted scent identification evidence to be too facile. It is no answer to say that "there is little distinction between a scent lineup and a situation where a dog is required to track" ( Winston v. State, supra, 78 S.W.3d at p. 527) and that it is "common knowledge that some dogs, when properly trained and handled, can discriminate between human odors." ( State v. Roscoe, supra, 700 P.2d at pp. 1319 1320, fn. omitted.) The trial court here recognized the significance of this distinction when it declined to permit a proffered defense expert to testify because his sole experience was with tracking and trailing, commenting that "dog trailing is a lot different than dog scent recognition." And while agreeing "that no one knows exactly how or why some dogs are able to track or scent, or the degree to which they are able to do so" ( State v. Roscoe, supra, 700 P.2d at p. 1319), we cannot ignore the California foundational requirement that scent identification evidence have a tendency in reason to prove a disputed fact. (See People v. Leahy, supra, 8 Cal.4th at pp. 597 598.)

Difficulty in understanding the precise nature and parameters of a dog's ability to discriminate scents does not take this phenomenon out of the realm of science. The feats attributed to Reilly, both in training and in the scent lineup conducted in this case, are truly extraordinary. An object may have many scents. The scent on an object touched by a target person will include the scent of any other person who touched the object. Here, for example, the prosecution asserted that at least two persons had touched the murder victim's shirt (Mitchell and the victim himself), and it may be that more than one person touched the bullets that were fired at the scene. Thus, to make a match, Reilly was required to compare the scent of every person who had touched the shirt with not only the scent of every person who touched the bullets, but also the scent of every person who had touched the detectives' chairs.

We do not suggest that Reilly's feats are beyond belief. Canine ability to discriminate scent has a physiological, and therefore in its most fundamental sense, a scientific basis. "The dog's nose maximizes the area available for exposure of chemical receptor cells to the air. . . . [¶] The total number of olfactory cells varies with the dog's size but in all breeds the number is far greater than in man. Thus, the German Sheep Dog has 220 million such cells compared to man's five million. Laid out, these cells cover an area in the dog's head that is about the size of the skin on his body, compared to humans, whose equivalent area of exposure of olfactory cells to air is the size of a postage stamp. Moreover, almost one-eighth of the canine brain is devoted to olfaction which the human olfactory lobes comprise a much smaller percentage of total brain size." (Taslitz, Does the Cold Nose Know? The Unscientific Myth of the Dog Scent Lineup (1990) 42 Hastings L.J. 17, 43, fns. omitted.)

Nevertheless, we are concerned in this case with the possibility that the scent of the shell casings found at the scene of the shooting may have been affected by the heat and pressure of being fired from a gun, the passage of time between when the casings were purportedly touched by defendant Mitchell, the conditions under which the casings were stored, and collection of the casings' scents by the scent transfer unit. Dog handlers D'Allura and Hamm testified that a scent will remain on an object for two to four months after it has been touched and that Reilly had succeeded in lineups conducted with objects that had been burned beyond recognition or surgically sterilized. But no effort was made to present information from any academic or scientific sources, let alone peer review journals, regarding these testimonial assertions. Thus, we are left with anecdotal rather than scientific explanations of Reilly's capabilities.

We are also concerned about the absence of any evidence that every person has a scent so unique that it provides an accurate basis for a scent identification lineup. Neither D'Allura nor Hamm, who also has no background in science, is aware of any scientific data which supports the notion that each person has a unique scent. On appeal, no such data has been brought to our attention.

If each person's scent is unique, it may be analogized to DNA. In contrast to the unsubstantiated claim of uniqueness here, the uniqueness of each person's DNA as expressed by the statistical probability of a random match between two people was the subject of extensive litigation under Kelly. (See People v. Axell (1991) 235 Cal.App.3d 836, 868 [method of calculating statistical probabilities generally accepted in scientific community]; People v. Barney (1992) 8 Cal.App.4th 798, 820 821 [debate in scientific community following Axell undermined the statistical method]; People v. Venegas, supra, 18 Cal.4th at pp. 82 90 [defining a statistical method that has become generally accepted]; People v. Soto (1999) 21 Cal.4th 512, 515 516 [addressing a variation on that method].) As stated in People v. Barney, supra, 8 Cal.App.4th at page 817, "A threshold issue is whether the Kelly-Frye requirement of general scientific acceptance applies at all to the statistical calculation step of DNA analysis. If not, the current scientific debate would go only to the weight of DNA evidence, not its admissibility, and would be a matter for jury consideration. [¶] . . . `[S]ince a match between two DNA samples means little without data on probability, the calculation of statistical probability is an integral part of the process and the underlying method of arriving at that calculation must pass muster under Kelly/Frye.' [Citation.]"

We further note that none of the dogs identified by breed in any of the tracking, trailing or scent identification cases that we have encountered is, like Reilly, a Labrador retriever. There are two reasons why this appears pertinent. First, foundational requirements for qualification of tracking and trailing evidence often include the element that the dog be of a breed, stock or pedigree characterized by acute powers of scent and discrimination. (See Annot., supra, 81 A.L.R.5th at p. 578; Winston v. State, supra, 78 S.W.3d at p. 526.) The record is devoid of any evidence to indicate that a Labrador is such a dog. Second, we observe the existence of controversy among handlers regarding the capabilities of certain breeds. While the Texas court sang the praises of the bloodhounds used in the scent lineup there ( Winston v. State, supra, 78 S.W.3d at pp. 526 527), in the case at bench expert witness D'Allura stated: "Bloodhounds, to be straight with you, they are pretty stupid. They just follow human scent. . . . They are not good at scent discrimination work at all."

Questions also come to mind regarding certification procedures for scent identification lineups. D'Allura testified that Reilly was certified by CARDA for search and rescue and searching for cadavers. Hamm, who is also affiliated with CARDA, testified that the bloodhounds he handles are utilized for tracking and trailing. We observe that CARDA, whose services are utilized by the California Office of Emergency Services (see Gov. Code, § 19844.5), has established requirements for certification in area searches, cadaver searches, and trailing, but not for scent identification. (CARDA Bylaws Standardshttp://www.carda.org/ppContents.aspx (as of July 3, 2003).) D'Allura testified that Reilly had a sheriff's department certification for article searches and also referred to a certification in scent identification lineups. But the record is silent as to any requirements of this certification.

Given the lack of scientific evidence, we are left in the dark as to whether the scent identification certification that Reilly received addresses the following concern. Here, scent taken from the evidence found at the scene (the shell casings) (1) was subjected to the potentially degrading environmental factor of being fired from a gun and (2) had come in direct contact with human skin (scent left by a person's fingers as contrasted with scent left on detectives' chairs). Do such distinctions in the origin of scents make any difference? Can a Labrador retriever, or any other dog, be trained to conduct a lineup in which such scents are involved? One commentator responded that "without scientific testimony establishing that dogs can indeed be so trained, the court or the jury cannot determine whether a dog's training was adequate." (Taslitz, supra, 42 Hastings L.J. at p. 121, fn. omitted.)

In ruling that Kelly did not apply to the scent lineup evidence and rejecting Mitchell's argument that an adequate foundation had not been established, the trial court essentially agreed with the approach taken in Winston v. State, supra, 78 S.W.3d at page 527, that the foundational elements required for admission of tracking and trailing evidence, such as those set forth in People v. Malgren, supra, 139 Cal.App.3d at page 238, could be modified to apply to this case. Nonetheless, as noted in People v. Leahy, supra, 8 Cal.4th at page 606, "a technique may be deemed `scientific' for purposes of Kelly/Frye if `the unproven technique or procedure appears in both name and description to provide some definitive truth which the expert need only accurately recognize and relay to the jury.' [Citation.]" A scent identification by Reilly appears to provide a definitive truth, with Reilly being analogous to a machine that D'Allura (and only D'Allura) can calibrate and read. Thus, we conclude that Kelly should have been applied to this evidence.

We further conclude that, even if Kelly were not deemed to apply to scent identification evidence in general, a greater foundation than the one provided here is needed for its admission. In tracking and trailing, there is a history of canine performance which provides the basis for the fifth Malgren element — that this type of evidence will be admitted if it is shown that the dog was put on a fresh trail. For scent identification to be relevant, there must be some basis for assumptions made about degradation and contamination of scent, both before and during collection, as well as the uniqueness of each person's odor, beyond the mere experiences of one trainer and one dog. Accordingly, scent lineup identification evidence was improperly admitted in this case.

iii. Prejudice

Although we have concluded that evidence of Reilly's scent identification was admitted in error, we find no prejudice. We have carefully inspected the exhibits in this case. Photographs of Lazaro's front door and the adjacent sidewalk reveal that Lazaro was less than 20 feet from Mitchell and Rogers when he viewed them on the sidewalk in front of his house for three or four seconds. The sketch prepared by the police artist working in conjunction with Lazaro on the day of the murder (August 20) bears a remarkably close resemblance to the profile photograph of Mitchell in a white T-shirt from which Lazaro made his identification on September 12. Nor do we see any great significance in Lazaro's failure to identify positively the photograph of Mitchell in the six-pack shown to him on August 24 (Lazaro said that Mitchell "looks like" the person with the gun), inasmuch as that photograph depicts a then barely 16-year-old, when his features were less mature. Notwithstanding whether Lazaro might have been better able to identify the perpetrators had he been wearing his glasses when he looked out the screen door, we assess his identification of Mitchell as being quite strong.

In addition, fellow gang member Hartfield (admittedly with something to gain from his testimony) stated that Mitchell had bragged about shooting a Hoover gang member whom he and Rogers had caught "slipping." This term was utilized by Mitchell in the rap lyrics he admittedly wrote, describing the shooting by saying: "Caught off brand slipping. Put two in his brain." Bullets of the same brand and caliber used in the murder were found in the search of Mitchell's home. And at some point following his arrest, Mitchell appears to have acquired a tattoo designating himself as a "Hoover killer."

Mitchell's claim of prejudice is further undermined by the fact that Rogers was also convicted, notwithstanding that the scent lineup evidence did not incriminate him. Other than the scent lineups, Rogers was faced with the same type of evidence as Mitchell, primarily Lazaro's identification and Hartfield's testimony about Rogers and Mitchell bragging on the day of the murder. If anything, the case against Rogers was weaker than against Mitchell, given that Lazaro failed to identify Rogers from the six-pack array and made only a tentative identification when all were wearing the white T-shirts. By convicting Rogers, the jury apparently found Lazaro's and Hartfield's testimony not only credible but compelling beyond a reasonable doubt.

It thus follows that it is not reasonably probable Mitchell would have been acquitted absent admission of scent identification lineup evidence, and the error in admitting that evidence must be deemed harmless. ( People v. Watson (1956) 46 Cal.2d 818, 836.)

2. Photographic Identification Procedures

3. Bruton Issue

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4. Impeachment of Hartfield

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DISPOSITION

The judgments are affirmed.

We concur:

SPENCER, P. J.

ORTEGA, J.