AINSWORTH, Circuit Judge:

The district judge in a bench trial convicted defendant of robbery in violation of C.Z.Code tit. 6, § 2361. Defendant contends that the court erred in admitting the robbery victim’s pretrial identification of defendant, because the police lineup from which the identification was made was im-permissibly suggestive. Finding no error therein, we affirm.

On February 20, 1977, shortly after noon, Dagoberto Espinoza was walking on the Bridge of the Americas in the Canal Zone, carrying a camera. As he walked, Espinoza “had a feeling something was going to happen,” turned his head and observed for two or three seconds a pair of men crossing the street. Espinoza again observed the two men for three seconds when he turned his head and saw them jump a small fence in the road. Espinoza was then grabbed from behind, and his camera was taken from him. An on-duty policeman observed the robbery from distances of 75-120 feet, and his description of the event substantially corresponded to that of the victim.

Espinoza viewed a police lineup on April 5, 1977, one and a half months after the robbery. He examined the lineup for one or two minutes and then told the attending officer that “maybe” number one (defendant) was the assailant, but he was not sure. The officer asked Espinoza to take another look, Espinoza complied, and then concluded that defendant was in fact the man who robbed him. 1 When asked if he was certain of the accuracy of the identification, Espinoza indicated that he had no doubt whatever. Prospero Ellis, the police officer who witnessed the crime, also identified defendant in a police lineup on April 5,1977. Ellis testified that he picked defendant out of the lineup “right away.”

The testimony of Ellis and Espinoza regarding their lineup identifications of defendant and the police photograph of the lineup which Espinoza viewed were admitted at trial, without objection from the defense. Counsel for the defendant did not cross-examine Espinoza regarding the composition of the lineup. On appeal defendant seeks reversal on the basis that the lineup was impermissibly suggestive because defendant was much darker complexioned than the other participants. In evaluating this contention we must first inquire “whether the identification procedure was unnecessarily suggestive.” Allen v. Estelle, 5 Cir., 1978, 568 F.2d 1108, 1112. If the procedure was unduly suggestive,'we must determine if the “procedure created a substantial risk of misidentification.” Id.

*285 Examination of the photograph of the lineup reveals that defendant was darker than the other participants, but not the only participant with dark skin. Eight men took part in the lineup, and although it is difficult to discern shades of complexion in the unevenly illuminated black and white print, it appears that the participants’ complexions ranged from light brown to defendant’s dark brown. There is no evidence that either the victim or the eyewitness policeman described the assailant to the police so that the police knew that the perpetrator of the crime was very darkskinned.

As no objection was raised at trial, our review of the propriety of admitting the identifications is limited to determining if plain error was committed. E. g., United States v. Mireles, 5 Cir., 1978, 570 F.2d 1287. Although this court has recognized that the race and other physical characteristics of participants can be a factor in evaluating the fairness of a lineup, see Pearson v. United States, 5 Cir., 1968, 398 F.2d 684, 688, there is insufficient evidence for us to conclude that defendant’s darker complexion caused the lineup in this case to be unduly suggestive. 2

In Manson v. Brathwaite, 432 U.S. 98, 114, 97 S.Ct. 2243, 2253, 53 L.Ed.2d 140 (1977), the Supreme Court emphasized that “reliability is the linchpin in determining the admissibility of identification testimony.” Even if it had been demonstrated that there was a significant degree of suggestiveness in this identification lineup, the' indicia of reliability employed in Brath-waite are satisfied under the facts of this case. Brathwaite examined the five aspects of reliability listed in Neil v. Biggers, 409 U.S. 188, 199, 93 S.Ct. 375, 382, 34 L.Ed.2d 401 (1972). The first factor is the witness’ opportunity to view. Here, the victim observed his assailant twice, for two or three seconds on each occasion, before he was robbed. The second indicator is the degree of attention of the observer, which was high in this case, as Espinoza was suspicious and turned around twice in order to view the two men behind him. The third element, the accuracy of the description, is not present in this case, as no evidence of a prior description was offered. The witness’ level of certainty was also satisfactory in this case. Espinoza suspected that defendant was the perpetrator of the robbery after approximately two minutes of viewing the lineup, and after taking a second view stated that he “remembered very well.” The final determinant of reliability is the time between the crime and the confrontation. Here, one and a half months passed between the robbery and the lineup. We cannot say that one and a half months presents an insurmountable barrier to the reliability of this identification. 3

Thus, we conclude that the lineup was not shown to be unnecessarily suggestive. Further, the standard of reliability stated in Brathwaite has been met. Admission of the identifications was therefore not plain error.

AFFIRMED.

Notes

. Espinoza described the identification procedure as follows: “[the police officer] asked me which one of these men was maybe the one who attacked me. And I told him maybe number 1, but I wasn’t sure. So he told me to go back and take a good look at him and come out and tell him for sure if the man that attacked me was there.”

. In Caver v. Alabama, 5 Cir., 1976, 537 F.2d 1333, 1335, this court stated that “[i]n Foster v. California, 394 U.S. 440, 89 S.Ct. 1127, 22 L.Ed.2d 402 (1969), the Court held that a lineup is unduly suggestive when it is virtually inevitable that the witness will select the individual whom the police have singled out.” (emphasis added)

. In Neil v. Biggers, 409 U.S. 188, 201, 93 S.Ct. 375, 383, 34 L.Ed.2d 401 (1972), seven months elapsed between the crime and the identifica *286 tion. The Court did emphasize, however, special circumstances in that case militating in favor of reliability.