The La Habra branch of the Bank of America was robbed by two armed men on the afternoon of May 14, 1976. The first man entered the bank through the rear door and ordered the employees and customers to get on one side of the bank. Almost immediately another man entered the front door, leaped over the counter, grabbed some money, and jumped back over the counter. The robbers left the bank approximately two minutes after their entry, taking with them $17,255. The bank's automatic surveillance cameras photographed the robbery.

The appellant Barron, identified at trial as the front door robber, was convicted of armed bank robbery in the United States District Court for the Central District of California. On appeal Barron urges that his conviction must be reversed because of impermissibly suggestive pretrial identification procedures, exclusion of his alibi witnesses, and unwarranted restrictions on defense counsel's closing arguments. We have considered all of the arguments raised by the appellant, and we affirm.I. Pretrial Identification Procedures

Seven eye-witnesses to the robbery testified for the prosecution and identified Barron at trial as the front door robber. Their participation in various pretrial identification procedures was elicited on cross-examination. These procedures included a preindictment lineup,1 a 25-photo spread of surveillance pictures, a six-photo spread, and pictures of the lineup. Only two of the prosecution witnesses had attended the lineup, and both identified Barron at that time as well as at the trial. Most of the photographic identifications occurred within a few weeks after the lineup, with two exceptions. One of the lineup viewers was shown the six-picture spread a week before trial, and one witness made his identification before the lineup. The seven prosecution witnesses based their in-court identifications of Barron on their recollections of the robbery. During extensive cross-examination the defense counsel brought out some inconsistencies in the witnesses' testimony, but he was unable to diminish the certainty with which they made their identifications. Four other eye-witnesses testified for Barron. Despite the suggestiveness which the appellant claims tainted the lineup, these four witnesses had been unable to identify Barron at the lineup. However, one of them made a positive identification of Barron at trial, where she was able to see him better.

Barron points to several ways in which the pretrial identifications were suggestive: 1) he was the only participant in the lineup with a big nose; 2) only one of the 25 surveillance pictures displayed the victim bank in the background; 3) some witnesses viewed the allegedly suggestive surveillance photo-spread before viewing the pictures of the lineup, thus tainting their selections; 4) some witnesses who made identifications from the lineup pictures were aware that other witnesses had made a positive identification at the lineup; 5) one witness received non-verbal approval after his identification and another was told that she had made the correct selection.

In Simmons v. United States, 390 U.S. 377, 384, 88 S. Ct. 967, 971, 19 L. Ed. 2d 1247 (1968), the Supreme Court stated:

(W)e hold that each case must be considered on its own facts, and that convictions based on eyewitness identification at trial following a pretrial identification by photograph will be set aside on that ground only if the photographic identification procedure was so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.

Subsequent cases have made it clear that unnecessary suggestibility alone does not require exclusion of identification testimony. Manson v. Brathwaite, 432 U.S. 98, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977). "It is the likelihood of misidentification which violates a defendant's right to due process . . . ." Neil v. Biggers, 409 U.S. 188, 198, 93 S. Ct. 375, 381, 34 L. Ed. 2d 401 (1972). Hence, the focus is on the reliability of the witnesses' identification rather than on the flaws in the pretrial identification procedures. Although the Simmons test speaks in terms of a pretrial identification from photographs, the same standard applies to identifications which have been preceded by attendance at a lineup. Foster v. California, 394 U.S. 440, 89 S. Ct. 1127, 22 L. Ed. 2d 402 (1969).

Even if it is conceded that elements of unnecessary suggestibility exist, certain factors must be examined to determine the likelihood of misidentification. As set out in Neil v. Biggers, 409 U.S. at 199-200, 93 S. Ct. 375, these factors include the opportunity of the witness to view the criminal during the crime, the witness' degree of attention, the accuracy of the witness' prior description of the suspect, the witness' level of certainty at the prior confrontation, and the length of time between the crime and the pretrial identification. Application of these criteria to the facts of this case convinces us that the identifications at trial were reliable. The witnesses had approximately two minutes in which to observe the robbers in a well-lighted bank in daytime, and they testified that they had an excellent opportunity to see Barron during the course of the robbery. Three witnesses had seen him before he entered the bank, two of them as he stood outside immediately before the robbery and one of them a week earlier in the bank. Although the two minutes were tense, the witnesses stated that they looked directly at Barron. One witness said that he had been trained to observe closely in such a situation. Although some witnesses referred to their prior descriptions of Barron during the pretrial motion to suppress the identifications, Barron does not elaborate on this factor in his briefs. Although some inconsistencies occurred in the testimony, it does not appear that any of the prior descriptions were wholly inconsistent with Barron's appearance. Each witness expressed a high level of certainty about his or her identification. Although two months had elapsed between the robbery and the pretrial identifications, the Supreme Court has approved longer delays when other factors assured reliability. In Neil v. Biggers, supra, for example, seven months had elapsed, but the witness, a rape victim, had made no intervening identifications of any other suspect. There is no indication here that any witness has ever selected any other suspect. See also Coleman v. Alabama, 399 U.S. 1, 90 S. Ct. 1999, 26 L. Ed. 2d 387 (1970) (two months).

Barron finds some support for his argument that the lineup was unfair because of his large nose in Foster v. California, 394 U.S. 440, 89 S. Ct. 1127, 22 L. Ed. 2d 402 (1969), where the Supreme Court held that the lineup was unfair in part because the defendant was much taller than the other participants and was wearing a jacket similar to one worn during the crime. However, in addition to these suggestive elements, the Foster Court also relied upon the fact that the witness had been unable to identify the defendant at two of three confrontations and the suggestive elements appeared to have induced the identification. The same cannot be said here. Nor are we convinced that the size of a nose must be treated the same as height in a lineup. Participants in lineups are obviously going to differ in facial characteristics, and it would be unduly burdensome to require police officials to find five or six very similar individuals for a lineup. The difficulty becomes readily apparent when it is noted that some of the witnesses also emphasized Barron's beady eyes and oddly shaped lips as two other distinguishing characteristics.

As for Barron's other allegations of suggestiveness, we do not find any of them serious enough to undermine the reliability of the eight positive in-court identifications. There is no indication that any of the witnesses received prompting prior to their selections. The witnesses who identified Barron in the surveillance picture testified that it was Barron and not the bank in the background which induced their identifications. Whatever flaws inhered in the pretrial identification procedures, we are convinced that the procedures were not so impermissibly and unnecessarily suggestive as to have created a substantial likelihood of misidentification. United States v. Higginbotham, 539 F.2d 17 (9th Cir. 1976); United States v. Baxter, 492 F.2d 150 (9th Cir. 1973).

Although a public defender was appointed for Barron and represented him at his bail reduction hearing (July 22, 1976) and arraignment (August 9), Barron attempted to retain private counsel up until the day set for trial to begin. When the case was called for trial on September 7, the public defender moved for substitution of counsel and a continuance and stated that he was not prepared to proceed because Barron refused to confide in a government-paid attorney and would not disclose certain witnesses and his alibi defense to the public defender. Noting the lateness of the request, the court denied the motions. After the jury was impanelled, court was adjourned until the morning of September 9. Meanwhile, Barron confided his alibi defense to his attorney, who then responded on September 8 to the government's August 27 request for notice of alibi witnesses under Fed.R.Crim.P. 12.1.2

On September 9, after the court had again denied renewed motions for continuance and substitution of counsel, the court was informed that Barron intended to rely upon an alibi defense although his Rule 12.1 response had been one day late. While the ensuing discussion may not have resulted in an outright ruling that Barron would not be permitted to call alibi witnesses, the defense counsel was at least put on notice before the trial began that the court might choose to exercise its discretion under Rule 12.1(d) to exclude the alibi testimony. The court criticized Barron for failing to cooperate with his attorney on his own defense and stated that Barron, rather than his attorney, had been neglecting his case. The court laid upon Barron the responsibility for complying with the alibi disclosure rule and warned that the defendant could not be permitted to benefit from his own negligence. The court concluded by agreeing to give the attorney "a few moments," but advised him to "read 12.1 and then see whether the defendant, not you . . ., whether the defendant has complied with 12.1." The government then proceeded with its case in chief. The next day, shortly before the government rested, the trial court expressly ruled that the alibi witnesses were barred from testifying because of the defendant's untimely response to the government's request for notice.3

Barron's attorney argues that the trial court abused its discretion in imposing the exclusion sanction provided in Rule 12.1(d), that good cause justifying an exception under Rule 12.1(e) was shown, and that the timing of the court's ruling denied Barron a fair trial because he was unable to alter trial strategy at that point to muster a good defense. The most cogent argument in Barron's favor is the timing of the government's initial request for notice. Since it was made only eleven days before trial, the government had obviously not allowed itself much time to prepare to meet the testimony of any alibi witnesses, even if Barron had made a timely response. Arguably, then, even with the response coming one day late, the government might have been able to interview the witnesses, some of whom were already known to the government, and to prepare a rebuttal without delaying the trial. We also note that the timing of the request for notice would have made it difficult for the government to have complied with its reciprocal disclosure duties under Rule 12.1(b) unless the court had chosen to reduce the time limits or to postpone trial. Of course, the government had already supplied the defense with a list of prosecution witnesses who would identify Barron as the front door robber, and it is unlikely that the government would have wished to add any other rebuttal witnesses. Perhaps this disclosure of identification witnesses would have satisfied that portion of Rule 12.1(b) which requires the government to respond to the defendant's notice by "a written notice stating the names and addresses of the witnesses upon whom the government intends to rely to establish the defendant's presence at the scene of the alleged offense . . ."

In any event, we would be ruling on hypotheticals if we based a decision on that reasoning. For the fact remains that Barron's response was tardy and the question of the government's obligation to respond never arose.4 The rule clearly empowers a district court to exclude the testimony of alibi witnesses in this situation, and the parties agree that abuse of discretion is the appropriate standard of review.5 While we do not believe that the sanction could be properly exercised in every case, we hold that on the facts of this case the district court did not abuse its discretion in ruling as it did and that it did not violate Barron's due process rights by ruling when it did.

As the Supreme Court stated in Wardius v. Oregon, 412 U.S. 470, 473, 93 S. Ct. 2208, 2211, 37 L. Ed. 2d 82 (1973), notice-of-alibi rules

are based on the proposition that the ends of justice will best be served by a system of liberal discovery which gives both parties the maximum possible amount of information with which to prepare their cases and thereby reduces the possibility of surprise at trial.

The rule also serves to prevent trial delays which might otherwise be necessary to permit a surprised party to meet an unexpected witness or alibi defense. See Notes of Advisory Committee following Rule 12.1. If these reasons are to be effectuated and if the rule is to have any teeth, trial courts must be able to impose sanctions, even the drastic one employed in this case. However, courts should impose the sanction only after a careful weighing of the interest of the defendant in a full and fair trial against the interests of avoiding surprise and delays.

We are particularly concerned that courts be cautious in assuming that criminal defendants are aware of the technical requirements of procedural rules. Here, for example, it is unclear from the record whether Barron's attorney had advised him of the deadline for responding to the government's request for notice and whether Barron knew the possible sanction which he faced. However, Barron at least knew that his trial was set for September 7 and he had no right to expect a continuance to be granted on the very day his trial was scheduled to begin. When his attorney first moved for a continuance on September 7, he stated that Barron refused to disclose certain witnesses and his alibi defense. Thus, one factor justifying the district court's ruling in this case was Barron's steadfast refusal to cooperate in his own defense up until the scheduled trial date because he did not trust a government-paid attorney.

Another factor justifying the sanction here is the strength of the government's case. Eight witnesses, including one of the defense witnesses, made positive courtroom identifications of Barron. By the time the court ruled on the alibi defense, six prosecution witnesses had concluded their testimony. The strength of their identifications no doubt influenced the district court's decision to impose the exclusionary sanction. As the Supreme Court stated in Williams v. Florida, 399 U.S. 78, 81-82, 90 S. Ct. 1893, 1896, 26 L. Ed. 2d 446 (1970):

Given the ease with which an alibi can be fabricated, the State's interest in protecting itself against an eleventh-hour defense is both obvious and legitimate. . . . The adversary system of trial is hardly an end in itself; it is not yet a poker game in which players enjoy an absolute right always to conceal their cards until played.

While the sanction might be entirely improper in a weaker case for the prosecution we do not find it improper here. Although the appellant argues that no delay would have occurred because the government had ample time to interview the alibi witnesses, this argument is merely speculative. The Notes of the Advisory Committee recognize that even when the government knows the identity of defense witnesses, the government may still be surprised to learn that they intend to testify about an alibi.

The appellant argues that his right to a fair trial was particularly prejudiced by the timing of the court's ruling. It is not entirely clear how Barron's trial strategy would have differed if the ruling had come on September 9 rather than on September 10. Because of the late response to the government's request for notice, the court could not have ruled any earlier than September 9. At the very least, Barron and his attorney were clearly put on notice before the trial got underway that the alibi witnesses might be barred. Any changes in trial strategy, including the decision whether Barron should take the stand, should certainly have been under consideration from the first moment that the question was raised. As we noted earlier, the delay in the ruling permitted the court to assess the strength of the prosecution's case and to exercise its discretion accordingly.

During his closing argument, the defense attorney asked Barron to smile in order to reveal that his two upper front teeth were missing. The attorney then proceeded to argue that the identification witnesses had failed to mention this distinguishing physical characteristic. The court curtailed this line of argument after a bench conference with the attorney in which the court explained that the argument went beyond the evidence. The defense had presented no evidence of Barron's dental condition at the time of the robbery.

The appellant cites cases which permit comments upon the defendant's courtroom demeanor in support of his argument that the court erred in ruling out that portion of the closing argument. The appellant misses the point: The problem lay not in Barron's courtroom appearance or demeanor but in his appearance on the day of the robbery. We see no error in the court's ruling. United States v. Marshall, 532 F.2d 1279, 1286 (9th Cir. 1976).

AFFIRMED.