The opinion of the Court was delivered by
POLLOCK, J.The primary issue is whether, in the absence of statistical probability evidence, the trial court erred in admitting expert testimony concerning the similarity in composition of lead bullets found at the crime scene, in the victim’s body, and among defendant’s belongings. Finding that statistical evidence was essential, a majority in the Appellate Division reversed the conviction of *144defendant, Judel Noel, for purposeful or knowing murder, N.J.S.A 2C:ll-3a(l), and possession of a handgun for an unlawful purpose, N.J.S.A. 2C:39-4a. One judge dissented, reasoning that the absence of statistical evidence affected the weight, not the admissibility of the expert testimony. The State appealed as of right. R. 2:2-l(a)(2).
We reverse the judgment of the Appellate Division and reinstate the convictions. We hold that statistical probability evidence is not a prerequisite to the admission of expert testimony concerning the composition of lead bullets.
I.
As Antoine Hargrove was returning to his home in Newark, he was shot in the back. He died at University Hospital several hours later. Two bullets were recovered from his body. At the crime scene, police recovered six 9mm shell casings made by Speer, a cartridge manufacturer, and four spent bullets. Two witnesses saw defendant flee from the scene.
The police arrested defendant at a pre-parole halfway house. A search of defendant’s locker revealed a pouch containing eighteen 9mm bullets, nine manufactured by Speer.
At the request of the police, Charles Peters, a physical scientist with the materials analysis unit of the Federal Bureau of Investigation, examined fifteen bullets: four collected at the crime scene, two recovered from the decedent’s body, and the nine Speer bullets found among defendant’s personal belongings.
Peters analyzed the bullets using a process known as inductively coupled plasma atomic emission spectroscopy (ICP). ICP determines the proportions of six elements other than lead: copper, antimony, bismuth, arsenic, tin, and silver. The bullet manufacturer adds these elements to each batch of lead. From one batch to another, the proportions in bullets of the six elements vary. Thus, the chemical composition of a bullet from one batch may *145match that of another bullet from the same batch, but not the composition of a bullet from another batch.
Peters divided the bullets into five compositional groups. Within each group, the bullets were of the same composition. Four of the five groups contained both a bullet from defendant’s pouch and one recovered either from the crime scene or from the victim’s body. For example, Group One included six bullets that were analytically indistinguishable: one bullet from the crime scene, one from the victim’s body, and four from defendant’s pouch. Group Four, which consisted of a solitary bullet found at the crime scene, did not match any other bullets.
At trial, Peters testified that, in his experience and that of his unit, “bullets that come from the same box have the same composition of lead and bullets that come from different boxes ... will have different compositions.” He explained that the manufacturer fills a given box with bullets from a single batch of lead. Consequently, those bullets will possess the same chemical composition. Because mixing may occur during storage, however, bullets of different compositions may be found in the same box. Peters concluded that he would not expect random batches of lead to produce the match that existed among the subject bullets.
Before conducting his analysis, Peters had visited the Speer manufacturing plant in Lewiston, Idaho. He limited his testimony on the manufacturing process to an explanation that each bullet is extruded from a “billet,” or seventy-pound cylinder of lead. Each batch of lead produces a number of billets. A billet yields approximately 4300 bullets. About five billion bullets are manufactured in the United States each year, and at least fifty thousand bullets may have the same composition.
The Appellate Division found that the trial court had committed reversible error in allowing Peters to testify, absent foundation evidence of statistical probability, about the identical composition between the bullets recovered from the crime scene and the victim’s body and those found in defendant’s pouch. 303 N.J.Super. 435, 445, 697 A.2d 157 (App.Div.1997). As the Appellate *146Division perceived the issue, Peters’s testimony depended on the statistical probability that the two sets of bullets would have the same composition. Ibid. According to the dissent, however, the absence of a statistical foundation affected the weight, not the admissibility, of Peters’s testimony. Id. at 453, 697 A.2d 157. The dissent pointed out that Peters’s testimony was not that the bullets at the crime scene came from defendant’s bag, but that some of the bullets from the crime scene and defendant’s pouch came from the same batch. Id. at 458, 697 A.2d 157.
In addition, the Appellate Division divided on the issue of the influence exerted by the expert’s testimony. The majority believed that the expert’s “extensive and impressive credentials” resulted in an “unwarranted enhancement of probative weight.” Id. at 445, 448, 697 A.2d 157. The dissent, by contrast, noted defense counsel’s “probing and able cross-examination of the expert,” id. at 458, 697 A.2d 157, and concluded that the expert’s testimony “merely added another link to the chain of evidence,” id. at 455, 697 A.2d 157.
Historically, statistical evidence has not been a prerequisite to the admission of matching samples. For example, in cases involving matching blood samples, statistical evidence of the probability of a match has not been required to establish a blood stain as a link in the chain of evidence. State v. Beard, 16 N.J. 50, 58-59, 106 A.2d 265 (1954); State v. Kelly, 207 N.J.Super. 114, 121-22, 504 A.2d 37 (App.Div.1986). Similarly, expert testimony about matching soil and hair samples has been deemed admissible, with the weight of the evidence left to the jury. State v. Baldwin, 47 N.J. 379, 392, 221 A.2d 199 (1966). Finally, expert testimony about matching carpet fibers has been admitted in the absence of statistical evidence about the probability of the match. State v. Koedatich, 112 N.J. 225, 548 A.2d 939 (1988); State v. Hollander, 201 N.J.Super. 453, 467-68, 493 A.2d 563 (App.Div.1985).
In Koedatich, a capital ease, the State presented evidence of matching fibers from the defendant’s automobile carpet and seat covers. Koedatich, supra, 112 N.J. at 242, 548 A.2d 939. The *147defense attacked the weight of the evidence by showing that manufacturers produced hundreds of thousands of yards of such fibers in a given year. Id. at 245, 548 A.2d 939. We upheld the admission of the evidence of the matching fibers, observing that the quantity of the fibers went to the weight, not the admissibility, of the evidence.
Similarly, in the present case, the expert’s testimony established a match among the bullets found in defendant’s belongings, at the crime scene, and in the victim’s body. Defendant contends that the large quantity of bullets produced by the manufacturer renders the match among the bullets inconclusive. As with the matching fiber samples, however, the production of a large quantity of comparable samples affects the weight, not the admissibility, of the evidence.
In reversing defendant’s conviction because of the lack of statistical evidence regarding the incidence frequency and distribution of bullets, the Appellate Division relied on our decision in State v. Spann, 130 N.J. 484, 617 A.2d 247 (1993). Spann, however, is distinguishable.
In Spann, the State sought to prove that the defendant had sexually assaulted the victim, who subsequently gave birth to a child, through DNA analysis of the blood tissue of the defendant and the child. The State’s expert testified to a 96.55% likelihood that the defendant was the father of the child. Finding the testimony inadmissible, this Court reversed the conviction and remanded the matter for retrial. The expert’s opinion, which was presented as “scientific” and “objective,” relied on the assumption that the probability of paternity before the analysis was 50%. The prior probability of paternity was based on the assumption that it was as likely that the defendant was the father as it was that he was not. Stated numerically, the prior probability of paternity was 0.5. The flaw in the assumption is that the prior probability of paternity must vary with the facts of each case. Otherwise, the probability would not vary even if the defendant were out of the country at the time of conception. No one, however, informed the *148jury of the effect that a different estimate of probability would have on the calculation of the probability that the defendant was the father. Thus the jury was unable to calculate the probability of paternity even if, on considering facts other than the blood and tissue analysis, its estimate of the probability differed from that of the State’s expert. In that context, the expert testimony usurped the role of the jury and compelled a verdict of guilt.
Unlike in Spann, the jury in the present ease received the guidance it needed to discharge its function. The expert explained the chemistry of lead analysis. He also explained why bullets of the same chemical composition generally came from the same box and why a single box may contain several bullets of different compositions. Left for the jury was the determination whether the bullets at issue came from the same box.
In explaining to the jury the issue of the prior probability of paternity, the State’s expert in Spann relied on Bayes theorem, a mathematical concept used in probability analysis. By contrast, the jury in the present case could evaluate the expert’s testimony without recourse to mathematical calculations. Like juries assessing samples of blood, soil, and fibers, the jury here did not require statistical data to discharge its duties. Peters’s testimony was comparatively straightforward. Contrary to the Appellate Division, we conclude that his opinion as an expert was not likely to create an “unwarranted enhancement of probative weight.” 303 N.J.Super. at 445, 697 A.2d 157.
Our conclusion comports with that of courts from other jurisdictions. For example, the Federal Court of Appeals for the Eighth Circuit has held that questions regarding whether bullets come from the same box affect the weight of the evidence rather than its admissibility. United States v. Davis, 103 F.3d 660, 673-74 (8th Cir.1996), cert. denied, Davis v. United States, 520 U.S. 1258, 117 S.Ct. 2424, 138 L.Ed.2d 187 (1997). The court pointed out that “[vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evi*149dence.” Id. at 674 (citing Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 596, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)). Similarly, the Supreme Judicial Court of Massachusetts has allowed an FBI agent to testify that bullets in the victim’s body and those found on defendant “come from the same box of ammunition or from different boxes that were manufactured at the same place on or about the same date.” Commonwealth v. Daye, 411 Mass. 719, 587 N.E.2d 194, 207 (1992). Finally, the Supreme Court of Oregon permitted expert testimony that bullets could have come from the same batch of metal, noting that the defendant’s expert properly pointed out the weaknesses of the evidence. State v. Krummacher, 269 Or. 125, 523 P.2d 1009, 1012-13 (1974).
ICP is an accepted method of bullet lead analysis. The compositional match among the bullets increased the probability that the bullets in the victim came from the defendant. That evidence constituted a link in the prosecution’s chain of evidence. The defense attempted to undermine that conclusion by cross-examining the expert, by showing that many bullets of the same composition had been manufactured, and by arguing an alternative conclusion to the jury. Consequently, we find that the trial court did not err in permitting Peters to testify about the similarity of the composition of the lead bullets.
We also conclude that Peters did not exceed the limits of his expertise in testifying about the manufacturing process. Peters testified that bullets of the same composition generally come from the same box, although a single box may contain bullets of several different compositions. He based his testimony on years of analyzing boxes of bullets and on a tour of the Speer plant. That tour may not qualify him as an expert on bullet manufacturing for all purposes. When combined with his substantial experience in analyzing bullets, however, the tour provided him with the “minimal technical training and knowledge essential to the expression of a reliable opinion.” Hake v. Township of Manchester, 98 N.J. 302, 316, 486 A.2d 836 (1985); see Landrigan v. Celotex Corp., 127 N.J. 404, 421-22, 605 A.2d 1079 (1992) (permitting *150epidemiologist to testify that asbestos can cause colon cancer); Rubanick v. Witco Chemical Corp., 125 N.J. 421, 426, 452, 593 A.2d 733 (1991) (allowing biochemist to testify that PCBs can cause colon cancer). Although experts generally may not express opinions outside their areas of expertise, those areas may overlap, and in certain circumstances an expert in one area may be qualified to express an opinion in another. Rosenberg by Rosenberg v. Cahill, 99 N.J. 318, 331-34, 492 A.2d 371 (1985). Here, Peters’s testimony regarding the arrangement of bullets in a box provided an appropriate basis for the jury to evaluate the significance of the bullet matches.
Underlying our opinion is the rationale that jurors will draw the appropriate inferences from matching samples such as fibers, soil; blood, or bullets. Our holding does not preclude an objecting party from offering statistical evidence to rebut the relevance of such samples. The admission of statistical evidence, like that of matching samples, is a matter that initially reposes in the sound discretion of the trial court.
Our dissenting colleagues agree that “there need not be any stated percentage of probability before an expert witness may testify about the composition of lead bullets.” Post at 153, 723 A.2d at 607. The dissent asserts, however, that the need for such a statement “is not the thrust of the Appellate Division’s opinion.” Ibid. That assertion ignores the primary point of disagreement in the Appellate Division, which was the need for probabilistic testimony. See ante at 145-46, 723 A.2d at 604. Because of that disagreement, this Court heard the matter as an appeal as of right under R. 2:2-l(a)(2). In amicus briefs, moreover, the Attorney General, the Public Defender, and the Association of Criminal Defense Lawyers have debated the issue vigorously.
For our dissenting colleagues, the issue is whether Peters’s testimony provided an adequate basis to support the conclusion that “the bullets not only ‘came from the same source of lead at the manufacturer’ but were ‘sold from the same box.’ ” Post at *151153-54, 723 A.2d at 608. According to the dissent, the issue is not whether Peters’s testimony regarding the matches between the bullets was admissible, but whether too many bullets were in circulation “to justify any conclusive inference of guilt.” Post at 154, 723 A.2d at 607. This issue focuses on whether the evidence was sufficiently rehable to permit the jury to infer that the various bullets came from the same box. A second concern of the dissent is that the prosecutor’s summation elevated the testimony from “a bit of circumstantial evidence that adds to the State’s case” to scientific fact, led the jury to ignore the large number of bullets in circulation, and so prejudiced the jury that we must set aside its verdict. Post at 156-57, 723 A.2d at 609-10.
The dissent charges that “the State was able to present this case to the jury as though it had scientific proof that the bullets in question came from the same box, even though there were at least 49,985 other bullets in circulation.” Post at 154, 723 A.2d at 608. According to the dissent, “[t]he problem in the case is not with what the expert testified to, but with what the State has attempted to do with his testimony.” Post at 154, 723 A.2d at 608.
In particular, the dissent highlights three statements from the State’s summation:
It is a very precise scientific process____
You could almost see [Peters] in a white lab coat. You could almost see him in math class in high school in the back. He had all the answers. He’s a straight shooter. ...
The key ... is the number of sources of lead; the number of batches. Millions of batches; each one unique like a snow flake, like a fingerprint.
[Post at 155-56, 723 A.2d at 608-10.]
At trial, defendant did not object to the first two statements. Not even in the Appellate Division did he challenge them. In overruling defendant’s objection in the prosecutor’s final statement to the analogy between snowflakes and bullets, the trial court characterized the statement as a “metaphor.”
In his own closing argument, defense counsel, apparently anticipating the prosecutor’s summation, argued that many boxes con*152tain bullets matching the ones at issue. That argument directed the jury’s attention to the issue that concerns the dissent, “whether too many bullets were in circulation to justify any conclusive inference of guilt.” Post at 154, 723 A.2d at 608. During the course of the trial, moreover, defense counsel vigorously cross-examined Peters. Finally, nothing prevented defense counsel from introducing evidence contradicting Peters’s testimony or from requesting a charge on the jury’s use of that testimony if it found the evidence to be unreliable or misleading.
Peters did not testify about the probability that the bullets came from defendant’s bag. Contrary to the dissent, moreover, his testimony did not constitute prejudicial scientific testimony that the bullets came from the same box. His testimony merely showed that some of the bullets from the crime scene, defendant’s bag, and the victim’s body contained the same trace elements. As such, the testimony constituted a link in the chain of evidence connecting defendant to the murder.
Excessive statements from both sides are a regrettable fact of life in criminal trials. In such trials, an objection by counsel remains as the first line of defense. Although the prosecutor’s statement may have been more temperate, it, particularly in the absence of an objection, does not justify upsetting the jury verdict. Given the realities of adversary proceedings, the prosecutor’s remarks pass as fair comment.
The judgment of the Appellate Division is reversed, and defendant’s conviction is reinstated.