On June 3, 1993, a federal grand jury returned an indictment in the district court for the District of Maryland, charging the Appellant, Douglas Fred Dorsey, with the May 19, 1993 robbery of the Bank of Baltimore. A superseding indictment was returned on July 8, 1993 adding two additional counts charging Dorsey with the March 31, 1993 robbery of the Custom Savings Bank (Count 1), the April 26, 1993 robbery of a Signet Bank (Count 2), and the May 19, 1993 robbery of the Bank of Baltimore (Count 3).

On July 21, 1993, Dorsey pled not guilty to the superseding indictment. Prior to trial, he filed a motion to suppress evidence seized during the search of his apartment, vehicle, and work locker; he also filed a motion to suppress pretrial identifications by witnesses, and a motion to sever the counts. The motions were denied.

Dorsey's trial on Counts Two and Three of the superseding indictment began on November 1, 1993, the government having elected not to proceed with Count One. When the case went to the jury, a redacted superseding indictment was submitted which renumbered the two remaining counts as Count One (Signet Bank robbery) and Count Two (Bank of Baltimore robbery). Dorsey was convicted on Count Two, and acquitted on Count One.

On November 16, 1993, Dorsey filed a motion for a new trial based on allegations that the district court had erroneously excluded certain expert testimony, had erroneously admitted evidence regarding the identity of Dorsey, and had erroneously failed to declare a mistrial. The motion was denied by the district court.

On January 13, 1994, Dorsey was sentenced to a term of 215 months in prison, with three years of supervised release. Dorsey has appealed his conviction on three separate grounds.

The two robberies at issue1 occurred in the following manner. The Signet Bank robbery occurred on April 26, 1993. At the time of the robbery, the bank teller, Sharon Keeley, was working at teller station 3 located in the center of the bank. Keeley described the robber as a black male, wearing a sweatshirt tied over his head and a construction vest with orange striping, and carrying a pair of work boots tied together by their laces over his shoulder. The robber caught her attention because he was overdressed on what was a warm day. He dropped a note on Keeley's counter which read: "This is a holdup--no 50s, no 100s, no dye pack or I will shoot you." He then reached over her counter, scooped up $559 and his note, and fled. Keeley testified that she could see the robber's face, but could not see his hair or ears because he was wearing a hood. Although he was wearing sunglasses, Keeley claimed she could see his eyes. The robbery was photographed by the bank surveillance cameras.

Approximately three weeks later, on May 19, 1993, a robbery occurred at the Bank of Baltimore. Betty Habersack, the victim teller, described the robber as a black male, wearing coveralls, a hat, glasses, a dust mask covering part of his face, and work boots laced together and hanging off of his shoulder. Unlike in the Signet Bank robbery, the robber attempted to obscure his appearance by wearing a surgical mask that covered a small portion of his face. As in the prior case, his appearance caught the teller's attention because he was overdressed on what was a warm day.

In her testimony at trial, Habersack stated that because the other tellers were busy, she asked if she could assist the robber. The robber approached the teller station and presented a demand note which stated, "Give me all your money, no dye packs, or I will shoot." Habersack removed $980 cash, including five twenty dollar bait bills, and handed them to the robber. The robber took back his note. The robbery was recorded by the bank surveillance cameras, and Habersack testified that she had ample time to observe the defendant.

As the robber was leaving the Bank of Baltimore, he was spotted by Raymond Eibner, an off-duty Baltimore City police officer. Eibner testified that he noticed that the robber was wearing overalls, a surgical mask, and boots around his neck. Eibner followed the robber and observed him getting into a red car, either a Mustang or a Datsun. Subsequent to the Bank of Baltimore robbery, FBI agents determined that Dorsey was the registered owner of the red Ford Mustang described by Eibner. As described by Eibner, the Mustang had a missing front license tag.

On May 20, 1993, a photo array containing photographs of Dorsey and of five other black males was shown by the FBI to Habersack. Initially, she was unable to decide which of two of the six photographs portrayed the robber, at which point Special Agent Lane Betts asked her if viewing the bank surveillance photographs would refresh her recollection. After indicating that it would, Habersack identified Dorsey as the man who robbed her.

On the same day, the photo spread was separately shown to Keeley. After viewing the surveillance pictures, Keeley too identified Dorsey as the man who robbed the Signet Bank.

On Saturday, May 22, 1993, FBI agents arrested Dorsey and impounded his vehicle. Special Agent Henry Hanburger testified that he moved the car to the FBI headquarters garage and conducted a search for any contraband. During a comprehensive search of the vehicle on May 25, 1993, pursuant to Dorsey's written consent, Special Agent Betts discovered a folder in the trunk containing Dorsey's household bills and cash, including two of the twenty dollar bait bills stolen from Habersack's teller drawer. Additionally, during a search of Dorsey's residence pursuant to a search warrant, the agents found a pair of work boots with the laces still tied together. The boots were confirmed to be the ones seen during the robbery by both Habersack and Eibner.

At trial, both victim tellers made positive in-court identifications of Dorsey as the man who robbed them. Introduced for the jury's consideration were both the photo arrays shown to the tellers, and numerous surveillance photographs depicting each of the two robberies in progress.

In addition to the identifications made by the tellers, other evidence was presented at trial linking Dorsey to the Bank of Baltimore robbery. For example, the prosecution introduced the two twenty dollar bait bills found in the trunk of Dorsey's red Mustang and the pair of brown work boots with the laces tied together that were seized from his apartment only three days after the robbery. The evidence of Dorsey's involvement in the Signet Bank robbery consisted primarily of eyewitness testimony of the tellers, due to the lapse of time between that robbery and Dorsey's arrest.

At trial, Dorsey presented a defense of mistaken identity. As part of that defense, Dorsey sought to introduce the testimony of two forensic anthropologists who would testify that Dorsey was not the individual depicted in the Bank of Baltimore surveillance photographs. Notice of Dorsey's intention to introduce the testimony of the two anthropologists, however, was not provided to the government until the first day of trial. The district court refused to admit the evidence on the grounds that the experts would not assist the trier of fact and may in fact usurp the jury's role.

At the conclusion of trial, the jury convicted Dorsey of the robbery of the Bank of Baltimore, but acquitted him of the robbery of the Signet Bank. On November 18, 1993, Dorsey filed a motion for a new trial, which was denied by the district court. On January 13, 1994, Dorsey was sentenced on Count Two for a term of 215 months, with three years supervised release. In the instant appeal, Dorsey has challenged his conviction on three separate grounds.

Dorsey first contends that the district court committed reversible error by excluding the testimony of two defense witnesses in the field of forensic anthropology who were to have testified about their analyses of the bank surveillance photographs from the Signet Bank and Bank of Baltimore robberies. In particular, Dorsey contends that the district court erred in excluding such evidence because the crux of his defense turned on the theory of mistaken identity. We find that under an abuse of discretion standard, the district court did not err.

As part of the defense's presentation of its theory of mistaken identity, Spencer Jay Turkel and James Vandigriff Taylor, both forensic anthropologists, were hired to compare the surveillance photographs of the bank robberies to recent photographs of Dorsey and photographs of the boots which were seized from Dorsey's house. Notice of Dorsey's intention to rely upon the testimony of those experts, however, was first provided to the government by letter not until Monday, November 1, 1993, the first day of trial. It was not until the second day of trial that the government was provided with a report by the two experts, dated November 2, 1993, summarizing their conclusions. According to the report, the experts had been contacted on Friday, October 29, 1993 to serve as experts for the defense. The report concluded that the person depicted in the Bank of Baltimore surveillance videos was not Dorsey. The report was unable, however, to come to any conclusion regarding the Signet Bank videos.

At trial, the government attorney objected to the testimony of the two experts on the grounds that their testimony would "absolutely subsume the jury's function." In support of the government's objection to the introduction of the testimony of the two experts, the government attorney pointed out that the report's conclusions were based on an "awful lot of ifs and hypotheticals that one must believe in order to reach that conclusion," that the experts were not able to rule out Dorsey as the robber in the Signet Bank robbery, and that the photographs of the work boots were too unclear to reach any accurate conclusion. Additionally, government counsel reasserted that "to have an expert come in and reach that conclusion would basically take over the jury's function in the case and for that reason we would ask that the testimony be excluded."

The district court ruled to exclude the evidence. In particular, it stated:

I am not so sure this is a recognized science such as a forensic chemist, or forensic scientist who does fingerprints, who does chemical analyses, who does handwriting, they are recognized. I think ... what we are doing here is comparing, is comparing some photographs. What we are really asking this expert to do is to tell the jury not to believe the witnesses in this case, because the witnesses in this case have already made their identification of the same evidence. They have said I looked at the photographs at the bank and I have been able to I.D. these photographs that belong to Mr. Dorsey. And I think that becomes clearly a jury function as to whether they are or are not. They believe them, why should we need an expert to say that they are wrong? I don't believe an expert can usurp the jury function in that regard.... I don't believe that I would need it. He said he would conclude with a reasonable degree of scientific certainty. I don't even believe that is enough.

Additionally, the district court, in excluding the evidence, emphasized that the government had not been sufficiently warned in advance of the defense's anticipated use of the expert testimony, and thus was not afforded a meaningful opportunity to obtain their own expert.

The Federal Rules of Evidence ("FRE") clearly suggest that the district court did not abuse its discretion in excluding the expert evidence in this case. In particular, Rule 702 provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

The Supreme Court, in construing Rule 702 in Daubert v. Merrell Dow Pharmaceuticals, Inc., --- U.S. ----, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), recently articulated the definitive threshold standard governing the admissibility of expert testimony. In particular, Daubert addressed whether a district court erred in excluding certain expert testimony regarding whether the drug Bendectin causes birth defects, on the grounds that the expert testimony was not sufficiently based on epidemiological evidence to establish causation. In holding that the proper standard for admissibility is governed by the FRE, rather than by the Frye2 test, the Supreme Court held that for expert testimony to be admitted, the method of analysis used by the expert need not have gained "general acceptance," as was required under the Frye test. Daubert, --- U.S. at ----, 113 S. Ct. at 2794. In thus lowering the standard for admissibility of expert evidence, however, the Daubert Court found:

That the Frye test was displaced by the Rules of Evidence does not mean, however, that the Rules themselves place no limits on the admissibility of purportedly scientific evidence. Nor is the trial judge disabled from screening such evidence. To the contrary, under the Rules the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.

Id. at ---- - ----, 113 S. Ct. at 2794-95.

In so holding, the Daubert Court set forth a two-part test which must be met in order for such expert testimony to be properly admitted under the FRE: (1) the expert testimony must consist of "scientific knowledge"--that is, the testimony must be supported by appropriate validation; and (2) the evidence or testimony must "assist the trier of fact to understand the evidence or to determine a fact in issue." Id. at ----, 113 S.Ct. at 2795. In determining whether certain expert evidence properly satisfies the first "scientific knowledge" prong of the two-part test, the Court held that trial courts may consider several factors: (1) whether the theory or technique used by the expert can be, and has been, tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error of the method used; and (4) the degree of the method's or conclusion's acceptance within the relevant scientific community. Id. at ---- - ----, 113 S. Ct. at 2796-97.

In determining whether the evidence meets the second prong of the two-part test--that is, whether the evidence will be helpful to the trier of fact--the Supreme Court warned that throughout an admissibility determination, a judge must be mindful of other evidentiary rules, such as FRE 403, which permits the exclusion of relevant evidence "if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury." Id. at ----, 113 S.Ct. at 2798. Specifically, the Court held:

Expert evidence can be both powerful and quite misleading because of the difficulty of evaluating it. Because of this risk, the judge in weighing possible prejudice against probative force under Rule 403 of the present rules exercise more control over experts than over lay witnesses.

Id. (citations omitted). In so holding, the Court concluded:

Conjectures that are probably wrong are of little use ... in the project of reaching a quick, final, and binding legal judgment--often of great consequence--about a particular set of events in the past. We recognize that in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations. That, nevertheless, is the balance that is struck by the Rules of Evidence designed not for the exhaustive search for cosmic understanding but for the particularized resolution of legal disputes.

Id. at ---- - ----, 113 S. Ct. at 2798-99.

In United States v. Bynum, 3 F.3d 769 (4th Cir.1993), cert. denied, --- U.S. ----, 114 S. Ct. 1105, 127 L. Ed. 2d 416 (1994), we reaffirmed the well-settled principle that, even under the Daubert analysis, a trial judge has a great deal of discretion in deciding whether to admit or exclude expert testimony. In Bynum, we specifically considered whether the district court's admission of certain government expert testimony was erroneous, where the expert chemists testified as to whether certain vials of crack came from the same batch. 3 F.3d at 772-73. In finding that the district court did not err in admitting the expert evidence under the Daubert test, we expressly noted:

The Daubert Court did make clear, though, that district courts need not admit all "scientific" evidence without any regard for its reliability. Trial judges may consider whether the particular opinion or technique has been subjected to peer review, what the known rate of error of the technique is, and recalling Frye, whether it enjoys "widespread acceptance" in the community. The [Supreme] Court emphasized that it was prescribing a "flexible" rule, one committed, as are most questions of admissibility of evidence, to the discretion of the district courts.

Id. at 773 (emphasis added).

Moreover, in United States v. Harris, 995 F.2d 532 (4th Cir.1993), decided just a few weeks before Daubert, we employed a similarly deferential standard for reviewing evidentiary determinations by a trial court and affirmed a district court's exclusion of expert evidence on bases relevant here. In Harris, we considered whether the district court erred in excluding a defense expert's testimony regarding the psychological limitations of an eyewitness identification. 995 F.2d at 534. In holding that the district court did not err in excluding that testimony under FRE 702, we expressly found that in determining whether a particular expert's testimony is sufficiently helpful to the trier of fact to warrant admission into trial, the district court should consider whether the testimony presented is simply reiterating facts already "within the common knowledge" of the jurors. Id. Under this standard, we found that the district court did not err in excluding the expert testimony because jurors "using common sense and their faculties of observation can judge the credibility of an eyewitness identification" without the help of an expert. Id. at 535.

Applying the rules to be derived from FRE 702 and the cases construing it, it is clear that the testimony to be presented by the two forensic anthropologists in the instant case did not plainly satisfy the first prong of Daubert--that is, that the evidence to be presented by the experts amounted to "scientific knowledge." Indeed, at no point in his brief to the court did Dorsey allege that the first prong of Daubert had been met; he instead merely alleged that the district court did not properly engage in the Daubert analysis at all in excluding the evidence. Because an appellate court can affirm a trial court's opinion on different grounds than those employed by the trial court, Dorsey's challenge to the district court's evidentiary ruling must fail. Brewster of Lynchburg, Inc. v. Dial Corp., 33 F.3d 355, 361 n. 3 (4th Cir.1994) ("We have consistently recognized that even though we disagree with the reasoning of the district court, we may affirm the result on different grounds if fully supported by the record.").

Indeed, in conducting the proper Daubert analysis, it is clear that the evidence to be presented by the forensic anthropologists did not in fact satisfy the first prong of the Daubert test--that is whether the evidence amounted to "scientific knowledge." The first relevant factor under the Daubert analysis--whether the method used by the experts can be, and has been, tested--was not satisfied in the instant case; indeed, Dorsey never contended anywhere in his brief, or during trial, that the forensic anthropologists' method of analysis had been tested. The second relevant factor under the Daubert analysis--whether the conclusions reached by the experts have been subject to peer review and publication--is likewise not satisfied in the instant case. The third Daubert factor--the known rate of error--is potentially very high in the instant case. Indeed, throughout their report to the district court, the forensic anthropologists placed several express conditions on their conclusions; for example, they stated that the accuracy of their conclusion that Dorsey was not the man in the Bank of Baltimore surveillance photographs depended expressly on whether "these images were taken at the same angle to the camera." Fourth, there is no indication that the type of evidence offered here has been widely accepted in the relevant scientific community; indeed, Dorsey's counsel indicated that such expert testimony had only been previously offered twice in similar cases.

Even if the evidence satisfied the first prong of Daubert, the second prong of Daubert--whether the testimony would be helpful to a trier of fact--was not clearly met in the instant case. Indeed, under the "common knowledge" rule articulated in Harris, there is no indication that the expert testimony was at all necessary in the instant case; as noted by the district court, the comparison of photographs is something that can sufficiently be done by the jury without help from an expert. See United States v. Brewer, 783 F.2d 841, 842 (9th Cir.1986), cert. denied, 479 U.S. 831, 107 S. Ct. 118, 93 L. Ed. 2d 64 (1986) (in a pre-Daubert case, testimony was not allowed of a forensic anthropologist who was to testify that the defendant might not be the robber in a surveillance picture, on the grounds that the jury could make that comparison on its own). As suggested in Harris, expert testimony can be properly excluded if it is introduced merely to cast doubt on the credibility of other eyewitnesses, since the evaluation of a witness's credibility is a determination usually within the jury's exclusive purview. Because in this case, the expert testimony was introduced largely to cast doubt on the positive identifications of Dorsey by both bank tellers involved in the robberies, the evidence was properly excluded.

Also, Dorsey's heavy reliance on our pre-Daubert decision in United States v. Sellers, 566 F.2d 884 (4th Cir.1977), is severely misplaced. In Sellers, it was held, in a bank robbery case, that the district court improperly excluded testimony proffered by a defense expert regarding photo comparisons because the district court allowed "essentially similar testimony by the government's expert." 566 F.2d at 885. Indeed, in Sellers, we expressly held:

Under different circumstances, Rule 403 might sustain the ruling of the district court. But this rule may not be utilized to exclude the otherwise admissible opinion of a party's expert on a critical issue, while allowing the opinion of his adversary's expert on the same issue. The discretion allowed by Rule 403 must be applied evenhandedly.

Id. at 886 (emphasis added). Because the district court in the instant case did not allow the government to present expert testimony while excluding the defense's expert testimony, Sellers is easily distinguishable.

Likewise, Dorsey's reliance on the Fifth Circuit's decision in United States v. Alexander, 816 F.2d 164 (5th Cir.1987), cert. denied, 493 U.S. 1069, 110 S. Ct. 1110, 107 L. Ed. 2d 1018 (1990), is similarly misplaced. In Alexander, the Fifth Circuit found that the district court erred in excluding certain expert testimony regarding the dimensions of a robbery defendant's head because "the only substantial evidence connecting [the defendant] to the robbery was [a] bank employee's identification of [the defendant's] driver's license photograph." 816 F.2d at 168-69. In the instant case, there was a great deal of both physical and testimonial evidence linking Dorsey to the robberies, thus precluding the need for expert testimony on the matter.

Even after the Daubert decision, otherwise admissible evidence may be properly excluded under Rule 403 if its probative value is substantially outweighed by the danger of unfair prejudice, confusion, or misleading the jury. Indeed, the Daubert Court specifically held:

Expert evidence can be both powerful and quite misleading because of the difficulty of evaluating it. Because of this risk, the judge in weighing possible prejudice against probative force under Rule 403 of the present rules exercise more control over experts than over lay witnesses.

--- U.S. at ----, 113 S. Ct. at 2798 (citations omitted). Because in the instant case, the district court was concerned that the expert testimony would confuse and mislead the jury, the district court did not abuse its discretion in excluding the testimony.

Finally, the case law is clear that it is not an abuse of discretion for a trial court to disallow expert testimony where a late proffer of evidence by the defense substantially prejudices the government in its ability to find its own expert and conduct similar testing. In United States v. Curry, 977 F.2d 1042, 1052 (7th Cir.1992), cert. denied, --- U.S. ----, 113 S. Ct. 1357, 122 L. Ed. 2d 737 (1993), for example, the Seventh Circuit, in holding that the district court did not err in excluding expert testimony regarding the reliability of eyewitness identifications, found, among other things, that the district court did not abuse its discretion because the defendants gave the government only four days' notice of their intent to call their witness. See also United States v. Dowling, 855 F.2d 114, 118 (3d Cir.1988) (finding that there was no abuse of discretion in excluding expert testimony where the defendant gave the government only 5 days notice of the proposed testimony because the government "could not reasonably be expected to search for its own expert and find one available to come to the Virgin Islands in time to be given the available facts and the opportunity to assimilate them"). It is clear that Dorsey's notice to the government only on the first day of trial that he was going to call expert witnesses to testify is certainly a formidable reason in itself for the district court to have excluded the proffered expert testimony.

Second, Dorsey contends that the district court abused its discretion in failing to grant a mistrial where, in response to the defense counsel's questioning, a law enforcement officer made a brief reference to Dorsey's criminal history. We find that the district court's actions did not amount to a clear abuse of discretion.

During the defense's case, Special Agent Lane Betts was called for additional questioning regarding his recovery of the bait bills from the trunk of Dorsey's red Mustang. Defense counsel started his examination by questioning Agent Betts about Dorsey's height and weight. The following colloquy took place:

Q: Well, he told you he was 190 pounds?

A: As I recall, I think he said he was 198 pounds, actually.

Q: And is this information basically confirmed by his Maryland driver's license?

A: Those are the questions that I asked him on May 22nd.

Q: But you also have his driver's license?

A: I don't have his driver's license in front of me.

Q: Did you ever check that against his driver's license?

A: At some point I am sure I checked it against his motor vehicle records and criminal history records.

Q: Do you recall, whether or not, there was any difference between those two things?

A: I wouldn't want to guess without checking the records.

No further reference was made to Dorsey's criminal records during the remainder of the trial, and the defense counsel made no immediate objection to Special Agent Betts's testimony. At the end of Agent Betts's testimony, however, defense counsel moved for a mistrial on the basis of Betts's reference to Dorsey's criminal history record. The district court denied the motion, stating that any potential prejudice could be cured by proper jury instructions, and that Betts's reference to Dorsey's criminal history record was in any case inadvertent and very brief. Ultimately, the district court did give curative instructions to the jury, in which the court stated:

I should also advise you that the defendant in this case is not on trial for any other crimes, or any other acts or conduct not alleged in the indictment. You will have a copy of the indictment when you go into the jury room. No other acts and no other charges is the defendant on trial for.

It is well settled that a grant or denial of a motion for mistrial is within the sound discretion of the trial court, and will not be overturned absent a clear abuse of that discretion. United States v. West, 877 F.2d 281, 287-88 (4th Cir.1989), cert. denied, 493 U.S. 959, 110 S. Ct. 377, 107 L. Ed. 2d 362 (1989). In order for the trial court's ruling to constitute such an abuse of discretion, the defendant must show prejudice; no prejudice exists, however, "if the jury could make individual guilt determinations by following the court's cautionary instructions." Id. at 288.

Dorsey has not made a sufficient showing of prejudice for several reasons. First, the jury ultimately found Dorsey guilty of the Bank of Baltimore robbery, but not of the Signet Bank robbery, suggesting that the reference to his criminal history did not cause so much prejudice to Dorsey that the jury was willing to convict him on all counts. Indeed, in West, we noted that one of the most important factors indicating the absence of prejudice in that case was that the jury acquitted the defendant of at least some of the counts against him. The West opinion stated:

The jury's verdict indicates that it conscientiously followed these instructions and disregarded the improper statements and conduct of [the defendants]. [The defendants] were both found not guilty of the most serious charge.... In addition, the jury acquitted [one defendant] on eight counts, [and the other defendant] on seven counts ... It can be inferred that the jury carefully considered the evidence against each defendant and based its verdict solely upon that evidence.

877 F.2d at 288. Dorsey's acquittal on at least one of the two counts against him is therefore strong evidence that the district court's denial of his motion for mistrial did not result in the type of prejudice that warrants reversal.

Second, Dorsey's attempted reliance on United States v. Johnson, 610 F.2d 194 (4th Cir.1979), cert. denied, 446 U.S. 911, 100 S. Ct. 1840, 64 L. Ed. 2d 264 (1980), actually supports the district court's denial of his motion for mistrial. In Johnson, we addressed whether one defendant's testimony regarding another co-defendant's past crimes warranted a mistrial. In finding that there was insufficient prejudice to warrant reversal, it was found:

Whether prejudicial error has been committed must be determined on the basis of the record in its entirety and the result will generally turn on the facts of each case. The record before us discloses that ... viewed in the context of the entire trial, the strong curative instruction of the district judge was sufficient to dissipate whatever small amount of prejudice may have been created by the reference to the Georgia robbery. While we have reversed convictions in cases where evidence of other crimes had been improperly presented, in those cases the inadmissible evidence was not only prejudicial, but had been purposely introduced by the prosecution.... Absent such misconduct on the part of the Government counsel, the courts generally have discerned no reversible error where the trial court has acted promptly in sustaining an objection and advising the jury to disregard the testimony.

610 F.2d at 196-97 (emphasis added). It is, therefore, clear that Dorsey's challenge to the district court's denial of his motion for mistrial is unavailing because like in Johnson, the introduction of the inadmissible evidence in the instant case was not done purposely by the government attorney, and thus did not amount to governmental misconduct of the type that warrants mistrial. Indeed, Special Agent Betts made the statement about Dorsey's criminal record when he was testifying in response to questions by defense counsel, not by government counsel, and no plotting by the government that he sneak in the improper testimony was shown. There is no evidence supporting Dorsey's contention that Agent Betts "deliberately" interjected the reference into the testimony. See United States v. Vogt, 910 F.2d 1184, 1193 (4th Cir.1990), cert. denied, 498 U.S. 1083, 111 S. Ct. 955, 112 L. Ed. 2d 1043 (1991) (finding that testimony suggesting that the defendant had been previously jailed, in light of curative instructions, was not sufficiently prejudicial to warrant mistrial where "the matter only came up incidentally, neither the witness nor the prosecution made any repeated reference to it, and the court carefully instructed as to its permissible use").

Third, in light of the overwhelming evidence linking Dorsey to the Bank of Baltimore robbery, including several eyewitness statements and a great deal of physical evidence, it is highly probable that any prejudice to Dorsey was harmless.

Fourth, the district court's curative instructions properly informed the jury of the types of evidence that they should and should not consider in reaching its verdict, without unduly emphasizing the existence of any criminal history record. Moreover, had the district court specifically tailored a jury instruction telling the jury to disregard Betts's reference to Dorsey's criminal history record, the district court would have drawn undue attention to that evidence and may have caused more prejudice to Dorsey. In fact, the defense counsel, during trial, expressed that very concern to the district court, stating:

Well, your Honor, may we at least know what the instruction is going to be ahead of time, because I am not entirely sure that won't draw more attention to it.

For Dorsey now to argue that the jury instruction should have been more specifically tailored to the situation contradicts his earlier stance.

IV. "Presumption of Truthfulness" Instruction

Finally, Dorsey contends that the district court committed plain error in instructing the jury that a witness is presumed to speak the truth. The instructions at issue were given as follows:

You, as jurors, are the sole judges of the credibility of the witnesses and the weight their testimony deserves. Ordinarily, it is assumed that a witness will speak the truth. But this assumes [sic] may be dispelled by the appearance and conduct of the witness, or by the manner in which the witness testifies, or by the character of the testimony given.

You should carefully scrutinize all the testimony given, the circumstances under which each witness has testified, and every matter in evidence which tends to indicate whether a witness is worthy of belief. Consider each witness's intelligence, motive, and state of mind, and demeanor and manner while on the stand. Consider also any relation each witness might have to or be affected by the verdict and the extent to which, if at all, each witness is either supported or contradicted by other evidence in the case.... Credibility is not merely choosing between one witness or another. As to each witness you are free to reject all that testimony, accept all that testimony, or as a third alternative reject some part and accept some other part of their testimony.

Although "presumption of truthfulness" instructions often, when not corrected, amount to reversible error under a harmless error standard, they do not constitute plain error. Thus, Dorsey's last challenge to his conviction is unavailing.3

Dorsey has placed a great deal of reliance on United States v. Varner, 748 F.2d 925 (4th Cir.1984), a case in which we found that a "presumption of truthfulness" instruction amounted to reversible error under a harmless error standard. Varner specifically held:

This court has held that the instruction that a witness is presumed or assumed to tell the truth is improper. "The jurors are the sole judges of the credibility of the witnesses and the weight to be given their testimony. This important function should not be encumbered by an assumption that witnesses speak the truth."

748 F.2d at 927 (quoting United States v. Safley, 408 F.2d 603, 605 (4th Cir.1969)). Accordingly, we found that the instruction was erroneous under the harmless error standard.

We, however, have taken a much more deferential approach to the issue when such "presumption of truthfulness" instructions are challenged under a plain error standard, such as in the instant case.4 In United States v. Love, 767 F.2d 1052 (4th Cir.1985), cert. denied, 474 U.S. 1081, 106 S. Ct. 848, 88 L. Ed. 2d 890 (1986), for example, we found that a "presumption of truthfulness" instruction did not amount to plain error, holding:

This court has long held that the instruction that a witness is presumed or assumed to tell the truth is improper.... In this case, however, no objection to this part of the charge was made. This failure to object precludes our review of the challenged instruction unless it constitutes "plain error".... In United States v. Safley, we considered whether a "presumption of truthfulness" instruction such as the one tendered here constituted plain error. We held that it did not. Safley governs our disposition of this issue on appeal today. The trial court did not commit plain error by issuing this instruction.

767 F.2d at 1060.

Such a ruling is particularly appropriate for extension to the present case where the presumption of truth instruction was immediately followed by "but" clauses pointing to exceptions and emphasizing that credibility was entirely an issue for the jury. See also United States v. Safley, 408 F.2d 603, 605 (4th Cir.1969), cert. denied, 395 U.S. 983, 89 S. Ct. 2147, 23 L. Ed. 2d 772 (1969) (because the defendants did not object to the presumption of truthfulness instruction at trial, the instruction, in light of the full charge, did not amount to plain error).

AFFIRMED.