Appellant Torrence Henderson was found guilty of participating in the armed robbery of the United Missouri Bank South in violation of 18 U.S.C. Sec. 2113(a) and (d) (1982).1 He was sentenced to twenty-five years in prison. On appeal, he argues that the district court erred in denying his motion to suppress the in-court identification of him by witness Harold Shaffer which he claims was tainted by a suggestive pretrial photographic showup.2 We affirm.

Three armed masked men robbed the United Missouri Bank South about 5:40 p.m. on June 25, 1982. At approximately 5:00 p.m. that afternoon, two black males approached Harold Shaffer in his car and offered him five dollars if he would jump-start their car. Shaffer agreed and the three departed in his car ostensibly to locate the disabled vehicle. After a short ride, Shaffer was forced out of the automobile at gunpoint and the two men drove off. The car was used in the bank robbery and later abandoned.

Shaffer immediately reported the theft and described his assailants to the responding officer. He gave a second description at the police station two hours later. Following this second description, Shaffer looked at some mug books but was unable to make an identification. He next examined a photographic lineup containing five pictures, including one of Henderson. Again, no identification was made. Shaffer was then shown a single photograph of Henderson which he positively identified. Later that same evening, he identified Henderson for a second time in a lineup.

Henderson filed a pretrial motion to suppress all identifications made by Shaffer, including the anticipated in-court identification. In denying the motion, the court3 did not determine the admissibility of Shaffer's two out-of-court identifications of Henderson because the government had decided to rely exclusively on his anticipated in-court identification. In admitting the in-court identification, the court concluded that "[w]e cannot say as a matter of law that his anticipated in-court identification was tainted by the out-of-court identification procedures followed by the police." The court also made it clear that both cross-examination of Shaffer and final argument would provide Henderson's counsel with the opportunity to convince the jury that the accuracy of Shaffer's in-court identification was suspect in light of the suggestive photographic showup. At trial, Shaffer identified Henderson as one of the persons involved in the theft of his car. On cross-examination, Henderson's lawyer fully explored the circumstances surrounding the use of the photographic showup and the out-of-court identification.

Due process challenges to convictions based on in-court identifications which follow a suggestive out-of-court confrontation are reviewed under a two-step test. Manson v. Brathwaite, 432 U.S. 98, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977); Simmons v. United States, 390 U.S. 377, 88 S. Ct. 967, 19 L. Ed. 2d 1247 (1968); United States v. Manko, 694 F.2d 1125 (8th Cir.) cert. denied --- U.S. ----, 103 S. Ct. 1224, 75 L. Ed. 2d 460 (1983). The first step is to determine whether the challenged confrontation between the witness and the suspect was "impermissibly suggestive." Simmons, 390 U.S. at 384, 88 S.Ct. at 971. If so, the second inquiry is whether, under the totality of the circumstances of the case, the suggestive confrontation created "a very substantial likelihood of irreparable misidentification." Manson, 432 U.S. at 116, 97 S.Ct. at 2254. This test reflects the fact that not all impermissibly suggestive confrontations give rise to a very substantial likelihood of irreparable misidentification. Ruff v. Wyrick, 709 F.2d 1219 (8th Cir.1983) (per curiam); United States v. Love, 692 F.2d 1147 (8th Cir.1980). Those identifications which are reliable--where the witness's perception of the suspect unaided by the suggestive confrontation provided a sufficient foundation for the identification--are admissible. Reliability is determined by examiningthe opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of his prior description of the criminal, the level of certainty demonstrated at the confrontation, and the time between the crime and the confrontation. Against these factors is to be weighed the corrupting effect of the suggestive identification itself.

Manson, 432 U.S. at 114, 97 S.Ct. at 2253 (citing Neil v. Biggers, 409 U.S. 188 at 199-200, 93 S. Ct. 375 at 382, 34 L. Ed. 2d 401 (1972)).

The first step is to determine if the photographic showup was impermissibly suggestive. While showups are "the most suggestive, and therefore the most objectionable method of pre-trial identification," United States v. Cook, 464 F.2d 251, 253 (8th Cir.) (per curiam) cert. denied 409 U.S. 1011, 93 S. Ct. 457, 34 L. Ed. 2d 305 (1972), whether or not they are impermissibly suggestive depends on the circumstances surrounding their use.4

In the present case, the surrounding circumstances aggravated the inherent suggestiveness of the showup. Shaffer overheard police radio broadcasts which caused him to conclude that the persons who stole his car immediately used it to rob a bank. Once at the police station, he overheard that two women had acknowledged who the two suspects were. The police then told Shaffer that they were going to bring in the two suspects identified by the women and that he should remain at the station for a possible identification. Finally, Shaffer testified that the police asked him "was this one of the bank robbers, the man that took [your] car" when they presented Henderson's picture to him for identification. Trial Record (I) at 89. Given these events, Shaffer could not help but expect that the photographs he was about to examine were of the named bank robbery suspects, which he had already concluded were the same persons who stole his car. Combined with the showup's intrinsic suggestiveness, these events created an impermissibly suggestive confrontation. Simmons, supra; Styers v. Smith, 659 F.2d 293 (2d Cir.1981).

Despite the impermissibly suggestive showup, however, we conclude that Shaffer's in-court identification of Henderson was reliable. First, Shaffer had ample opportunity to view Henderson. Shaffer spoke with him and his companion face to face for one or two minutes prior to entering the car. Once in the car, Shaffer and Henderson were together in the front seat for five to ten minutes. Shaffer was wearing his glasses. It was five o'clock in the afternoon, thus providing adequate lighting, and Henderson did not conceal his features in any manner. Second, Shaffer focused at least a normal degree of attention on Henderson during this time.5 Shaffer was not a bystander or casual observer. Moreover, because he was unaware of Henderson's criminal intent until the very end of their journey, his perceptions were not "clouded by the excitement of the [crime]." Hadley, 671 F.2d at 1115. Third, Shaffer's descriptions of Henderson were sufficiently accurate. He overestimated Henderson's height by 2-4 inches, his weight by 25-35 pounds, and underestimated his age by 4-8 years. We do not consider these deviations substantial. Shaffer did fail to mention Henderson's moustache and beard in both descriptions to the police. While this is a significant omission, it is not weighty enough to undercut the overall reliability of Shaffer's identification. Fourth, all three of Shaffer's identifications of Henderson--the showup, the lineup and in court--were certain and without mistake. Fifth, only hours separated the theft of the automobile and the showup. This short span of time decreases the likelihood that the witness is remembering the "person in the photograph more readily than the appearance of the person who committed the crime." United States v. Dailey, 524 F.2d 911, 914 (8th Cir.1975) (photograph shown to witness a month and a half after the crime).

Appellant claims that the district court's statement that a motion to suppress the out-of-court identification alone "would likely have been granted" necessarily implies that the district court believed that the out-of-court identification was unreliable, i.e., it resulted from an impermissibly suggestive confrontation which created a very substantial likelihood of irreparable misidentification. From this conclusion, appellant argues that the subsequent in-court identification must also be suppressed because it was the product of an inadmissible identification and cannot be any more reliable than it. By admitting the in-court identification, appellant claims that the district court misapplied the Manson standard.

We do not agree with appellant's reading of the district court's order. The court's ruling was based on the government's commitment to rely solely on the in-court identification, and the statement cited by the appellant, read in this light, is dictum. Second, in spite of its language, the district court did not conclude that the out-of-court identification created a very substantial likelihood of irreparable misidentification. The statement contains no findings that the out-of-court identification was unreliable, but was only an expression of the court's inclination which was not fully developed. We cannot conclude that such a statement is a bar to the testimony of Shaffer in open court which identified Henderson. The evidence was for the jury to weigh.

In balancing the reliability of the in-court identification with the "corrupting effect" of the photographic showup, we conclude that the witness possessed a foundation for the identification independent of the suggestiveness of the photographic showup. The conviction is affirmed.