(dissenting). Because the showup in this case was unwarranted and broadened the use of a showup in violation of this Court’s jurisprudence, I dissent. I would reverse the conviction.

The chronology of the relevant events is not in dispute. At 11:30 a.m. on July 6, 1999, two uniformed officers, Brian Holtje and Thomas Bafimdo, riding in a patrol car, received a radio report of a burglary at 51 Mills Pond Road in Smithtown, New York. About 20 minutes later, at 11:55 a.m., the officers arrived at the crime scene where they met with another officer, John Crowley, who had arrived earlier. Crowley told the two officers *598that the complainant, an elderly lady, saw a man fleeing her home upon her arrival. The complainant described the man as a shirtless 18- to 20-year-old white male with brown hair and muscular build wearing red or maroon (i.e., dark red) shorts.

The two officers then drove north looking for suspects. The officers stopped in front of 66 Mills Pond, which appeared to be under renovation. While Holtje knocked on the front door, which was open, Bafundo went around the back where he noticed a swimming pool, and a man inside the house wrapping himself in a towel. Bafundo then came back to the front, told Holtje that he saw a man inside the house, and then went back to the back of the house again. Holtje knocked on the door once more, and defendant answered, wearing a towel. Defendant told Holtje that the house belonged to his sister and that he was renovating the bathroom. To Holtje, defendant appeared to be 30 years old although he was actually 40. Also in the house was a white male in his twenties fully clothed with either brown or black hair.

Once Holtje and Bafundo were back in their car, Holtje expressed skepticism at defendant’s claim that the bathroom was being renovated, considering that he was wearing a towel. At that time, Holtje was not aware that there was a pool in the back. Holtje and Bafundo then drove around and contacted detective, Brian McNeil, who later met them in front of 66 Mills Pond. When Holtje knocked on the door, defendant answered, this time wearing tan shorts and no shirt. Defendant agreed to invite the officers inside. Once inside, Holtje went into a nearby bedroom where he noticed wet maroon shorts on the floor. Defendant stated that the shorts belonged to him. McNeil eventually asked defendant if he would go to the crime scene, and he agreed.

Another officer, Christine Ward, arrived in a patrol car and drove Holtje and defendant to the victim’s house, which was two minutes away by car. McNeil and Bafundo went in separate cars. Once they all arrived, around 12:25 p.m., defendant exited the police car and stood in the driveway, about 15 to 20 feet from the front of the house. He was wearing the tan shorts, no shirt, and he was not handcuffed. McNeil asked defendant to hold the shorts, and then went inside the house to meet the complainant. Apparently, defendant held the shorts “down to his side” “next to his hip.” Bafundo and Ward were standing beside defendant and Holtje was behind him. There were three vehicles in front of the house. As McNeil and the complainant were standing in the front room of the house looking out the *599window, McNeil asked her if she recognized anyone standing outside. According to McNeil, the complainant stated that defendant was “the person that she saw leaving the house, and that he was the same height, color hair, build, and she also identified the shorts that he was holding.”

Defendant was then asked to go to the precinct, but he was not arrested until three days later. At a suppression hearing, the trial court found that “(1) the show-up was conducted promptly, within a short time after the commission of the crime; (2) it was conducted at the crime scene; (3) defendant was not singled out — in fact, he was not even in handcuffs; and (4) he was allowed to leave after the victim saw him.” After his arrest, defendant was charged with burglary in the second degree and petit larceny. He pleaded guilty to one count of second-degree attempted burglary after receiving assurances that he could appeal the suppression ruling.

A majority of the Appellate Division affirmed, holding that the identification was proper. Also relevant, the Court found that the “show up was conducted in close temporal and geographical proximity to the crime scene.” (292 AD2d 626.) The majority rejected the argument that requiring defendant to hold the shorts during the identification was improper. The lone dissenter argued that “[t]here was no reason for him to hold the shorts, other than to single him out as the perpetrator.” (292 AD2d at 629.)

On appeal, defendant focuses on the argument of the dissent that the showup was unduly suggestive because he was required to hold the shorts, although he still argues that “no exigency existed.”

It has been said repeatedly that a showup — the presentation of a single witness for identification — is inherently suggestive and for that reason strongly disfavored (see People v Riley, 70 NY2d 523, 528 [1987]). That showup identifications are inherently suggestive means that they are likely to result in the identification of an innocent person as the perpetrator of a crime. Despite their inherent suggestiveness, showup identifications “are permissible if exigent circumstances require immediate identification (People v Rivera, 22 NY2d 453), or if the suspects are captured at or near the crime scene and can be viewed by the witness immediately (People v Love, 57 NY2d 1023)” (id.). Showups that are “not the product of police suggestion but rather spontaneous” or “the result of happenstance” need not satisfy temporal and geographic conditions (People v Clark, 85 NY2d 886, 888 [1995]).

*600In light of the difficulty of having to show that exigent circumstances compelled a showup of a defendant sitting in jail, rather than the less suggestive lineup, it should come as no surprise that precinct showups are presumptively infirm (Riley, 70 NY2d at 529). We have found exigent circumstances, however, where the eyewitness was in the hospital suffering from critical wounds (see Rivera). While street showups of suspects caught in or near the crime scene are not presumptively infirm, they “must be scrutinized very carefully for unacceptable suggestiveness and unreliability” (People v Duuvon, 77 NY2d 541, 542, 543 [1991]). We have stated that “[w]hile the limits of an appropriate time period between the alleged crime and a showup identification may vary from case to case, the emphasis must be upon the prompt and immediate nature of an identification after the crime has been committed * * *” (People v Johnson, 81 NY2d 828, 831 [1993]).

The cases where we have found no infirmity with the show-ups have generally involved a temporal span of 15 minutes or less between the crime and the showup. In some of these cases, the central focus was the legality of the stop or arrest. For example, in People v Brnja (50 NY2d 366 [1980]), the defendant and his accomplice were sitting in a van parked less than a mile from the crime scene when the police, acting on a description by the robbery victim and witness, handcuffed and frisked them, and then transported them back to where the crime had occurred 15 minutes earlier. We found that the arrest was based on probable cause and that the showup was valid “in view of its proximity in time and location to the point of arrest” (id. at 372). In People v Hicks (68 NY2d 234 [1986]), we found that the police had reasonable suspicion to detain and transport defendant and his accomplice to the crime scene, which was one minute away by car (id. at 240). In addition to defendants’ fitting the general description of the perpetrators, there were no other cars at the intersection where they were stopped, and defendant and the accomplice gave an answer regarding their prior whereabouts that the police knew was incorrect (id. at 237). Defendants were detained for about 10 minutes and the time between the crime and the identification was not much longer (id.).

In other cases, the legality of the showup was the central focus. In Duuvon, the defendant alone robbed a dry cleaners that he, as part of a group, had robbed 10 days earlier. As defendant made his way toward his getaway car, a taxicab, two employees who now had been twice robbed, and were deter*601mined to prevent a third, bellowed for help. Defendant got into the cab, but a police car blocked its path. Defendant then fled, but was caught seconds later, clutching in his hand a wad of cash. One of the employees immediately appeared and spontaneously identified the defendant. Defendant was arrested and taken back to the cleaners where he was identified by the other employee while in the back of the police car in handcuffs. Although we disapproved the use of the handcuffs, we found that the showup in front of the employee was “within the permissible boundaries of the governing legal principles” because it took place several minutes after the crime and around the corner from the scene (77 NY2d at 544). In addition, there was an “unbroken chain of events — crime, escape, apprehension and identifications * * *” (id. at 545).

As in this case, the complainant in People v Love (57 NY2d 1023 [1982]) came home to find a burglar. Unlike this case, however, a transit officer who immediately responded to complainant’s call for help saw the defendant on the fire escape of the building, chased him down the street, and upon apprehending him, brought him back to the building where the complainant identified him in the presence of bystanders. The showup took place five minutes after the crime.

In contrast to the foregoing cases, we found in Johnson that the showup was improper and that a lineup should have been conducted. There, the complainant was robbed by a perpetrator who then escaped. After the police arrived, they drove the victim around the neighborhood without finding the perpetrator. About 2V2 hours later, the police apprehended defendant near the crime scene and brought him back there. The police also drove the complainant, who was not nearby, to the crime scene where he identified the defendant.

Here, the identification took place about 55 minutes after the occurrence of the crime. Looking solely at the time span, 55 minutes exceeds by 40 minutes the longest temporal span we have permitted {see Brnja). While such a delay does not render the showup automatically invalid, it is an important factor militating against a finding that the show up was prompt and immediate, which could be rebutted by the circumstances of the case. But even when the time span is coupled with the circumstances of this case, the showup was not prompt and immediate. This is not a case that involved a fast-paced investigation. The officers did not see defendant run from the house, nor did they chase him down the street. On a very warm day, they found him at a nearby house with a pool, also oc*602cupied by another person, initially wearing a towel and then tan shorts. The officers knew where to find him, and had no trouble locating him the second time. That there was no urgent need for a showup is also evidenced by the fact that police did not arrest him until three days later.

Defendant was the only person the police suspected, even after their second drive around the neighborhood, and they obviously did not feel that they had sufficient information to detain or arrest him. This is significant because immediate showups (the one here was not immediate), are tolerated based “on [the] objective that the police have reasonable assurances that they arrested or detained the right person” (Duuvon, 77 NY2d at 545). In light of these circumstances, and the inherent suggestiveness of showups, a lineup should have been conducted.

Moreover, the showup was unduly suggestive. While show-ups are inherently suggestive, they become unduly suggestive when they create a “substantial likelihood of misidentification” (id. at 544). Here, looking through the window of her home, the complainant saw standing 15 to 20 feet away a man of the same height, hair color, and build as the one she had seen earlier, although much older, who was wearing and holding maroon and tan shorts, respectively, while flanked by three uniformed officers. In the background, there were two patrol cars and one unmarked car. The holding of the shorts “at his side,” even though they were his, while being virtually surrounded by officers, essentially rendered defendant a bull’s eye, and created a substantial likelihood of misidentification (cf. Riley, 70 NY2d at 527-528 [in a precinct showup, gun and stolen property were laid out on a table near the suspects who were the only nonuniformed persons in the room]).

Courts should do more than pay lip service to the acknowledgment that “ ‘[t]he influence of improper suggestion upon identifying witnesses probably accounts for more miscarriages of justice than any other factor — perhaps it is responsible for more such errors than all other factors combined’ * * *” (id. at 530 [citations omitted]). An identification plays a central role in the truth finding process of the criminal justice system. Since showups tend to undermine that process, they should be, in substance and in form, strictly scrutinized and admitted in exceptional circumstances. This is particularly the case given that “[t]he inadmissibility of the showup identification evidence alone * * * does not preclude admission of identifications subsequent to the showup ones if they, in turn, are justified by independent source reliability standards and if they are otherwise authorized” (id. at 531).

*603In summary, because the showup was unnecessary, untimely and suggestive, I dissent.

Chief Judge Kaye and Judges Ciparick, Wesley, Rosenblatt, Graffeo and Read concur; Judge Smith dissents and votes to reverse in an opinion.

Order affirmed in a memorandum.

OPINION OF THE COURT

Memorandum.

The order of the Appellate Division should be affirmed.

*597The sole issue on this appeal is whether the crime scene showup identification of defendant was unduly suggestive when defendant, who was wearing tan shorts and no shirt, was asked to hold a pair of maroon shorts. The maroon shorts belonged to defendant, were found at the house where he was located, and matched the identifying victim’s prior description of the clothing worn by the perpetrator.

We have allowed showup identifications, in the absence of exigent circumstances, where the showup was reasonable under the circumstances — that is, when conducted in close geographic and temporal proximity to the crime — and the procedure used was not unduly suggestive (see People v Ortiz, 90 NY2d 533, 537 [1997]; People v Johnson, 81 NY2d 828, 831 [1993]). Whether a crime scene showup is unduly suggestive is a mixed question of law and fact. Thus, if record evidence supports the determination below, this Court’s review is at an end.

Here, record evidence supports the conclusion that the procedures used were reasonable under the circumstances. The showup took place at the scene of the crime, within an hour of the commission of the crime, and in the context of a continuous, ongoing investigation.* Record evidence also supports the conclusion that the showup identification was not unduly suggestive. The victim stated that defendant was the person whom she had seen leaving her house, and initially and independently identified him relying on his height, hair color, and build. In these circumstances, the presence of defendant’s maroon shorts, admittedly his own, did not, as a matter of law, negate the reasonableness of the police action.