The defendant has been convicted after a jury trial of two counts of criminal sale of a controlled substance in the third degree and two counts of criminal possession of a controlled substance in the third degree, and sentenced to a concurrent term of 5 Vi to 11 years on each count. The principal evidence implicating him in the commission of these crimes was the testimony of undercover Officer Louis Prince. Prince testified at trial that he had purchased cocaine from the defendant on two occasions. The theory of the defense was that Prince had misidentified the defendant.
Central to this appeal is the defendant’s contention that Prince’s identification of the defendant at a station-house viewing some five hours after the defendant’s arrest ought to have been suppressed because the People did not meet their burden of going forward at the pretrial Wade hearing held to determine whether the station-house identification was sufficiently reliable to be admitted at trial.
The single witness at the Wade hearing was Officer Prince. Prince testified that he had made two purchases of narcotics from a person whom he referred to as J.D. Chief and another person who was never apprehended. The first of these purchases took place on March 26, 1987 at an apartment located at 505 West 164th Street in Manhattan. Apart from the fact *292that the transaction was brief, lasting about three minutes, there were no other details provided at the hearing illuminating the circumstances under which it was conducted. The second purchase took place on April 2, 1987 at the same apartment. Again, the purchase took only three minutes. During this time Prince noted that there were numerous people in the apartment. The person whom he referred to as J.D. Chief was the only black man he saw, but Prince acknowledged that he did not look in all of the apartment’s rooms. Shortly after the April 2, 1987 purchase, a warrant authorizing the search of the apartment was executed and those present in the apartment were arrested. No identification procedure was conducted immediately after the arrest. Rather, several hours later Prince was asked to come to the precinct to make an identification "because”, he was told, "they have some subjects from the warrant location”. Prince accordingly went to the precinct and some five hours after the drug purchase viewed a lineup through a one-way mirror. Of the people in the lineup Prince could recognize only the defendant whom he identified as J.D. Chief. Prince could remember nothing about those whom he had viewed along with the defendant, nor was there any photographic record introduced documenting the composition of the lineup.
In reports completed by Prince after the March 26 transaction and after the identification procedure conducted on April 2, Prince described J.D. Chief as a black man about 40 years old and 6-feet tall weighing approximately 150 pounds. Prince noted that J.D. Chief had a salt and pepper afro and that he had a mustache but made no mention of any other facial hair. Defendant’s arrest photograph, however, disclosed that he had a full beard and it was stipulated that at the time of his arrest he weighed 180 pounds.
In denying the motion to suppress the station-house identification, the motion court began by observing "[T]here is nothing on the face of it which suggests—which teaches me or which evidences any deliberate or even negligently suggestive procedure on the part of the police department.” The court then noted that Officer Prince was trained in making identifications and concluded that under all the circumstances his identification possessed "inherent reliability” and should not be suppressed.
It is clear that at a pretrial Wade hearing it is the People’s burden to come forward with evidence showing that the identification procedures used by the police were not violative *293of the defendant’s constitutional rights (see, People v James, 111 AD2d 254, affd on other grounds 67 NY2d 662; People v James, 110 AD2d 1037; People v De Congilio, 71 AD2d 990; People v Borges, 37 AD2d 581). This entails, at a minimum, a showing that the identifying witness initially observed the defendant under conditions which permitted the witness to commit the defendant’s appearance to memory and a further showing that the identification procedures subsequently employed were free from suggestive influences potentially leading to misidentification.
The People do not argue that they met this burden. Indeed, the argument would be tremendously difficult to make, for manifestly there was nothing in the record to support the motion court’s finding that the April 2 identification procedure had not been suggestive. Due to the People’s deficient presentation, the court knew nothing about the composition of the lineup in which the defendant had appeared and, therefore, was not in a position to draw any conclusion as to whether the lineup had been fair. The danger of effectively relieving the People of their burden was, of course, fully realized at trial when, with the station-house identification already before the jury, it was revealed that the lineup viewed by Prince had been composed entirely of those arrested at the 505 West 164th Street apartment and that not one of those persons resembled the defendant, even remotely; flanking the defendant in the lineup were a black woman, two hispanic women, and a black man who the People concede did not look like the defendant and who was, in any case, young enough to have been the defendant’s son.
As noted, it is not the People’s position that they carried the burden which would ordinarily be theirs at a Wade hearing, or even that they could have, for the lineup was by any standard patently suggestive. Rather, it is the People’s position that they had no burden at all because under the circumstances there was none but a negligible possibility that their witness would be influenced by even the most suggestive identification procedures. In this regard they stress that their witness was a trained undercover officer, that he had had two encounters with the defendant in the week preceding the defendant’s arrest, and that the second of these encounters occurred at "a relatively minimal” interval from the station-house identification and at a time when the officer had reason to believe that the defendant would shortly be arrested. In furtherance of this argument, that their witness was immune *294to suggestive influence and that his identification should, therefore, be exempt from the sort of scrutiny required at a Wade hearing in satisfaction of the due process guarantees of the Constitution, the People cite to the Court of Appeals memorandum in People v Wharton (74 NY2d 921) which they find dispositive.
In Wharton (supra) a police officer participating in what is called by law enforcement authorities a "buy and bust” operation, purchased cocaine from the defendant therein. Just after the purchase, the officer radioed a description of the defendant to his back-up team and the back-up team immediately proceeded to the cite of the transaction and arrested defendant. Within five minutes of the arrest the purchasing officer drove by the arrest site and identified the defendant as the person from whom he had purchased the cocaine. Some three hours later, the purchasing officer again identified the defendant, this time through a one-way mirror at the station house. The defendant’s motion for a Wade hearing was denied upon the ground that the station-house viewing was merely confirmatory. The propriety of that denial was subsequently upheld by the Appellate Division and by the Court of Appeals which stated in its affirming memorandum: "It is not disputed that the identification was made by a trained undercover officer who observed defendant during the face-to-face drug transaction knowing defendant would shortly be arrested. Thus, there is evidence in the record to support the determination of the courts below that the officer’s observation of defendant at the station house approximately three hours later was not of a kind ordinarily burdened or compromised by forbidden suggestiveness, warranting a lineup procedure or Wade hearing” (supra, at 922). The court in concluding, however, went out of its way to stress that this holding was not to be taken as a broad permission to dispense with Wade hearings, and to indicate that its affirmance of the Appellate Division’s order was dictated largely by its observance of its jurisdictional limitations: "In sum, we imply no categorical rule exempting from requested Wade hearings confirmatory identifications by police officers by merely labeling them as such (see, People v Perez, 74 NY2d 637). Where the nature and circumstances of the encounter and identification may warrant, a hearing should and undoubtedly will be held. However, there is no basis in this record for us to conclude that the trial court, as affirmed by the Appellate Division, erred as a matter of law in denying the suppression hearing and motion” (supra, at 923).
*295In the more recently decided case of People v Newball (76 NY2d 587) the Court of Appeals has reiterated that Wharton was not intended to create for police identifications a broad and ready exemption from scrutiny at pretrial Wade hearings (People v Newball, supra, at 592; see also, People v Gordon, 76 NY2d 595, 600-601). Rather, as the analysis in Newball demonstrates, each case must be carefully evaluated on its own facts.
The issue in Newball (supra) was whether the People were to be excused from giving notice, pursuant to the requirement of CPL 710.30, of their intention to introduce identification testimony at trial. In support of the dispensation sought, the People argued, as they do in this case, that because the identification of their police witness was merely confirmatory, it was not of a kind " 'ordinarily burdened or compromised by forbidden suggestiveness’ (People v Wharton, 74 NY2d, at 922, supra)” (People v Newball, supra, at 592). Thus, they argued that the mandated notice which would ordinarily trigger a defendant’s request for a Wade hearing where identity was at issue, was unnecessary because their police witness’s identification was so inherently reliable as to render pretrial scrutiny of the circumstances under which it was obtained, superfluous.
In rejecting this argument, the Newball court, citing People v Gissendanner (48 NY2d 543) noted that the only acceptable bases for dispensing with procedures mandated to assure the reliability of identifications to be used at trial, were that "the defendant’s identity [was] not in issue, or * * * the protagonists [were] known to one another” (People v Gissendanner, supra, at 552, cited in People v Newball, supra, at 591). And, in assessing whether the protagonists were in fact "known to one another”, the court found the following passage from People v Collins (60 NY2d 214) to be of particular relevance: "[Knowledge], of course, is necessarily a question of degree. When a crime has been committed by a family member, former friend or long-time acquaintance of a witness there is little or no risk [of taint through police suggestion]. * * * But in cases where the prior relationship is fleeting or distant it would be unrealistic to ignore the possibility that police suggestion may improperly influence the witness in making an identification” (supra, at 219, cited in People v Newball, supra, at 591).
As this is a case in which identity is most assuredly at issue, the predicate for the People’s claim that they were not re*296quired to go forward at defendant’s Wade hearing to establish that Officer Prince’s station-house identification of the defendant was not prompted by police suggestion, must be that Prince and the defendant were known to one another. Reduced to its essentials then the People’s central contention on this appeal is that two three-minute encounters with the J.D. Chief, the last of which occurred some five hours prior to the Prince’s purported identification of him, were sufficient to make J.D. Chief so well known to Prince that there was "little or no risk of taint through police suggestion.” This is simply not tenable.
In People v Wharton (supra) it will be recalled, the interval between the arrest and the drive-by identification by the undercover officer was a mere five minutes. The police having, by this procedure, already substantially eliminated the possibility that the wrong person had been taken into custody, cannot be said to have courted any but the most minimal risk of misidentification when, 3lá hours later the defendant was identified a second time by the undercover officer at the station-house showup. Here, by contrast, the first identification of the defendant occurred nearly five hours after Prince’s three-minute encounter with J.D. Chief. The notion that Prince’s memory of J.D. Chief must at the juncture have been accurate, much less indelibly so, is not supportable.
To begin, it is unarguable that Prince’s acquaintance with J.D. Chief did not even approach the level of familiarity that would ordinarily constitute a prima facie basis for dispensing with pretrial safeguards against mistaken identification testimony; Prince was not a member of J.D. Chiefs family, nor was he a former friend or long-time acquaintance (see, People v Collins, supra, at 219; People v Newball, supra, at 591). Indeed, Prince’s contact with J.D. Chief cannot be characterized as anything but fleeting. The People do not deny this but maintain nevertheless that they demonstrated at the Wade hearing that Prince’s station-house identification was merely confirmatory. In this connection they place the heaviest reliance on Prince’s training and the fact that he knew at the time of the April 2 transaction that J.D. Chief would likely be arrested in short order (see, People v Wharton, supra).
Although it may seem reasonable to believe that undercover narcotics agents are, in light of their training, possessed of superior powers of observation and identification, there is little in the way of objective evidence to justify this belief. Studies, in fact, have shown that as a general matter the *297identifications of police witnesses are no more reliable than those of laypersons (see, e.g., Tickner & Poulton, Watching for People and Actions, 18 Ergonomics 35 [1975]) and may even be less so (see, e.g., Krouse, Effects of Pose, Pose Change and Delay on Face Recognition, 66 J of App Psychology 651). In his dissent in People v Wharton, Judge Titone ably noted some of the reasons why police identifications ought not to be too readily relied upon:
"As cases such as People v Chillis [60 AD2d 968], People v Baez [103 AD2d 746], and People v Williams [79 AD2d 929] reflect, even trained police officers who have observed the suspect with professional detachment are subject to suggestive influences. Indeed, although the risk of misidentification may be diminished in cases involving police witnesses because of their expertise and relative detachment, there are special problems associated with station house police identifications following street 'buy and bust’ operations that raise reliability questions not present in ordinary pretrial identification procedures. As is evident from the trial record in this case, 'buy and bust’ teams ordinarily have many encounters and make many 'buys’ during the course of their shifts before they return to the station house to 'confirm’ the arrests made by their backup teams. Accordingly, they must retain in their memories the details of many intervening faces and events, all of which relate to essentially similar transactions. Furthermore, the undercover’s contact with the drug seller often is fleeting and takes place under circumstances that are not conducive to careful observation because of the officer’s wish to avoid arousing suspicion. Finally, although I do not suggest that a police officer would knowingly identify the 'wrong’ person, it cannot be denied that there is a certain degree of suggestion inherent in the very fact that the officer’s own backup team, consisting of individuals with whom he may work on a regular basis, has made an arrest on the basis of the officer’s description.
"For all of these reasons, I do not believe that the expertise and detachment that ordinarily accompanies identifications by police officers obviates the need to test such identifications for suggestive influences that might impair their accuracy. Accuracy of the postarrest station house identification is particularly critical in cases such as this, since, unlike the ordinary citizen witness who typically have few encounters with crime, a police officer may have contact with hundreds of suspects in the time between the arrest and the trial. Accordingly, the
*298police officer’s in-court identification testimony, if any, is likely to be discounted, and, concomitantly, his testimony describing the station house identification that took place shortly after the arrest is likely to take on great probative significance.” (People v Wharton, supra, at 929-930 [Titone J., dissenting]; see also, People v Gordon, supra, at 601.)
In the end, police training is but a factor to be considered in evaluating the reliability of an officer’s identification, it is not a guarantee of reliability, nor should an officer’s professional qualifications be relied upon to the point that a court’s inquiry into the circumstances of a particular identification is relaxed or, as here, practically abandoned.
Looking beyond the guarantee of reliability which the People falsely posit in Officer Prince’s training, it becomes apparent that Prince’s recollection of the defendant may well have been flawed and, at the very least, was not so certain as to have been impervious to suggestion. Given the People’s sparse presentation at the Wade hearing, which did not disclose in any detail the circumstances under which Prince’s observations of J.D. Chief were made, there was little basis to conclude that Prince had had the opportunity to form an accurate impression of J.D. Chiefs appearance. Assuming, however, that Prince had accurately described J.D. Chief in his buy reports, there were significant discrepancies between his description of J.D. Chief and the defendant. As noted, Prince made no mention in his reports concerning J.D. Chief of the full beard that was perhaps the defendant’s most prominent feature. And, in describing J.D. Chief as being 6-feet tall and weighing a mere 150 pounds he portrayed a man of slight build in distinction to the defendant who at the same height and weighing an additional 30 pounds would have possessed a notably heavier build. Thus, even if there were a basis to suppose that Prince had obtained a reliable impression of J.D. Chief, the question would still remain as to whether that impression retained its freshness and reliability some five hours after the protagonists’ encounter when Prince viewed the station-house lineup. It is not particularly helpful in answering this question that Prince had reason to believe at the time of the April 2 transaction that J.D. Chief might soon be arrested. Whatever Prince may have thought as to the imminence of the arrest, the fact remains that his identification was made considerably after the arrest when his memory of J.D. Chief may well have faded and become susceptible to suggestive influence.
*299Under these circumstances there was no basis to relieve the People of the burden, ordinarily theirs at a Wade hearing, to demonstrate that the identification procedures employed by the police were not suggestive. As the People failed to meet this burden, Officer Prince’s station-house identification of the defendant ought to have been suppressed.
Finally, there can be no claim, and indeed none is made, that the admission of Officer Prince’s station-house identification at the defendant’s trial was harmless. Prince’s identification testimony was the sole evidence implicating the defendant in the commission of the offenses for which he was convicted. Had Prince’s in-court identification been deprived of the bolstering effect of the much earlier and presumptively more probative lineup identification there might well have been a different result in this case (see, People v Dodt, 61 NY2d 408; see also, People v Wharton, supra, at 930 [Titone J., dissenting]).
Accordingly, the judgment of the Supreme Court, New York County (Jeffrey Atlas, J., at suppression hearing; Felice Shea, J., at jury trial and sentence), rendered August 29, 1988, convicting defendant after a jury trial of two counts of criminal sale of a controlled substance in the third degree, and two counts of criminal possession of a controlled substance in the third degree, and sentencing him to concurrent indeterminate terms of from 5 Vi to 11 years should be reversed and the matter remanded for a new trial, in connection with which the lineup identification of the defendant should be suppressed and a new Wade hearing held for the limited purpose of affording the People an opportunity to demonstrate that there was an independent source for Officer Prince’s in-court identification of the defendant.
Rosenberger, Ross and Asch, JJ., concur with Sullivan, J.; Murphy, P. J., dissents in a separate opinion.
Judgment, Supreme Court, New York County, rendered August 29, 1988, affirmed.