delivered the opinion of the court:
In December 1980, defendant, Steven Paul Linscott, was charged by indictment with three counts of murder (Ill. Rev. Stat. 1979, ch. 38, par. 9—1(a)(2)), two counts of armed violence (Ill. Rev. Stat. 1979, ch. 38, par. 33A— 2), and a single count of rape (Ill. Rev. Stat. 1979, ch. 38, par. 11—1(a)) in the death of Karen Ann Phillips. The armed violence "charges were dismissed before trial. A Cook County jury found defendant not guilty of rape but guilty of murder. He was sentenced to 40 years’ imprisonment.
This is the second time the case is before this court. Previously, the appellate court reversed defendant’s conviction, holding that the evidence of guilt was legally insufficient. (People v. Linscott (1985), 135 Ill. App. 3d 773.) This court granted the State’s petition for leave to appeal (103 Ill. 2d R. 315) and reversed the appellate court decision (People v. Linscott (1986), 114 Ill. 2d 340). The case was remanded to the appellate court for consideration of other issues raised on appeal that were not addressed by the appellate court. Linscott, 114 Ill. 2d at 349.
On remand, the appellate court again reversed defendant’s conviction and remanded for a new trial. (159 Ill. App. 3d 71.) The court held that defendant was denied a fair trial because, in closing argument, the prosecutor misrepresented two of the three elements of the State’s evidence against the defendant (i.e., the blood-typing and hair-comparison evidence). (159 Ill. App. 3d at 81.) We granted the State’s petition for leave to appeal (107 Ill. 2d R. 315).
The detailed facts of this case have been recounted in both appellate court opinions as well as in our earlier de*26cisión. Therefore, for purposes of this opinion, we will provide only a brief summary of the facts, highlighting those we consider relevant to our decision in this appeal.
Karen Ann Phillips, the victim, was found dead in her apartment in Oak Park on October 4, 1980. Police found the victim’s body face down and naked except for a nightgown pushed up around her neck and shoulders. An autopsy performed on the victim revealed that her death was the result of several blows to her head and strangulation.
Hairs were found, among other places, clasped in the victim’s hands, in her pubic region and in a carpet on the floor of her apartment. These hairs were removed and tests were conducted on them as part of the murder investigation. A vaginal swab was also taken from the victim and analyzed for blood type and the presence of seminal material.
On the afternoon of October 4, Oak Park police canvassed the neighborhood and spoke with defendant. At the time of the murder, defendant lived in the apartment building immediately next to that of the victim. The police asked defendant and his wife if they had heard or seen anything unusual earlier that morning. The police told defendant and his wife that someone had been murdered at about 1 a.m.
According to defendant, at that point he remembered a dream from the previous night involving a brutal beating. The dream occurred around 1 a.m. He did not mention the dream to the police at this time. However, on October 6, defendant called the Oak. Park police and related his dream to Officer Robert Scianna. Scianna asked defendant to produce a written version of the dream. Defendant did so and Scianna picked it up at defendant’s home the following day.
On October 8 and 10, defendant came to the Oak Park police station and gave two tape-recorded inter*27views describing the dream. After the second interview, defendant accompanied the officers to a hospital where he gave samples of blood, saliva and hair. Defendant and the police officers returned to the police station, where the officers accused defendant of committing the murder. Defendant was not arrested, however, until November 25. Defendant has at all times denied murdering the victim.
At trial, the State relied on three pieces of evidence to convict defendant of the victim’s murder: the dream defendant recounted to the police which, according to the State, paralleled the crime so closely that it was proof defendant committed the murder; head and pubic hairs found in the victim’s apartment and on her body which were consistent with defendant’s head and pubic hair; and the results of blood-typing tests which showed that the seminal material taken from the victim could have come from defendant.
This appeal concerns comments made by the prosecutor in argument with respect to the physical evidence in the case. (This case was tried by two prosecutors; “prosecutor” shall be used throughout this opinion when referring to one or both of the prosecutors.) The appellate court held that the prosecutor’s “closing arguments to the jury, relating to blood and hair comparisons, were so egregious that a denial of a fair trial resulted.” (159 Ill. App. 3d at 74.) It is unclear whether the appellate court reversed due to specific improper statements or the effect of all the improper statements. The court also concluded that “substantial rights” were affected by the improper comments because the evidence in the case is so closely balanced. (159 Ill. App. 3d at 74.) Thus, although no objections were made to the prosecutor’s comments at trial, the court applied the plain error doctrine to review the errors. (159 Ill. App. 3d at 74.) Based on this review, the appellate court remanded for a new trial.
*28We find that certain statements the prosecutor made during closing statements regarding the blood and hair comparisons were improper, and in addition, we find that statements the prosecutor made during opening were also improper. We also believe that application of the plain error doctrine is appropriate. (See, e.g., People v. Szabo (1986), 113 Ill. 2d 83, 94 (“Plain error *** is ***■ to be invoked only when the error alleged is ‘so substantial as to deprive defendant of a fair trial’ [citation]”); People v. Sanders (1983), 99 Ill. 2d 262, 273 (“The plain error doctrine is properly applied *** where the error alleged is so substantial as to reflect on the fairness or impartiality of the trial regardless of how closely balanced the evidence is [citations]”).) Because it is unclear whether the appelldte court reversed due to specific improper statements or the effect of all the improper statements, we will examine all of the statements the appellate opinion discusses.
Prosecutorial misconduct in closing argument warrants reversal of the conviction and a new trial if the improper “ ‘ “remarks *** constitute^] a material factor in the conviction.” ’ ” (People v. Lyles (1985), 106 Ill. 2d 373, 391, quoting People v. Fields (1974), 59 Ill. 2d 516, 522.) The issue is whether “the jury could *** have reached a contrary verdict had the improper remarks not been made.” (People v. Witted (1979), 79 Ill. App. 3d 156, 168.) If a reviewing court “cannot say that the prosecutor’s [improper] comments did not contribute to the defendant’s conviction[ ],” the court should order a new trial. Witted, 79 Ill. App. 3d at 166.
We will first examine the prosecutor’s comments with respect to the hair-comparison evidence. The appellate court concluded that the prosecutor erred by arguing in closing that hairs “matching” defendant’s had been found at the crime scene; by claiming that defendant’s hairs had been found there; by claiming that the State’s *29experts’ use of the word “consistent” meant the same thing as a layman’s use of the word “identical”; and by claiming that odds published by a forensic scientist, Barry Gaudette, were applicable to the hair comparisons made by one of the State’s experts. 159 Ill. App. 3d at 78-79.
Three witness testified on the subject of hair comparisons. Mark Stolorow, the coordinator of serology for the Illinois Department of Law Enforcement, testified for the State regarding the methods of hair comparison that are employed by the Department. Stolorow explained that through the use of a “comparison microscope,” a simultaneous visual comparison is made of the characteristics of hair samples from two different sources. This technique excludes classes of individuals from consideration as suspects in an investigation and is conclusive, if at all, only to negate identity. See 159 Ill. App. 3d at 78-79.
The State’s expert, Mohammad Tahir, a forensic scientist for the Illinois Department of Law Enforcement, testified as to the hair comparisons he performed on the hair samples from the victim and from defendant. Tahir testified that he looked at approximately 7 to 12 characteristics. Based upon the comparisons he performed, Tahir concluded that certain of the hairs found in the victim’s apartment were “consistent with” the samples provided by defendant. Tahir defined “consistent” as “no dissimilarity.” Tahir testified that defendant’s hair samples were “consistent” with those hairs found in the victim’s right hand, hairs found on the carpet, and two pubic hairs that were combed from the victim’s pubic region. On cross-examination, however, Tahir conceded that a person cannot be identified by the hairs he leaves behind:
“DEFENSE: And you sure can’t determine from whose head that hair came from, can you?
*30TAHIR: You cannot positively say.
* * *
Q. Okay, you couldn’t even say that if you had two pieces of hair from the same head, could you?
A. My answer is the same, what I told you that you cannot say that this hair came from this individual, only could say is that it is consistent with [sic].”
Despite this testimony, the prosecutor argued that hairs found in the victim’s apartment and on the victim’s body were in fact defendant’s hairs. In closing argument, the prosecutor told the jury that “the rug in the area where Karen was laying [sic] was ripped out sometime later, rolled up and shipped to the laboratory. And that another group of hairs were obtained. The head hairs of Steven Linscott.” In closing argument, the prosecutor made such claims repeatedly: “he [defendant] left eight to ten hairs of his in that apartment; “his [defendant’s] pubic hairs [were found] in her crotch”; and “his [defendant’s] hairs are found in the most private parts of the woman’s body.”
With these statements, the prosecutor improperly argued that the hairs removed from the victim’s apartment were conclusively identified as( coming from defendant’s head and pubic region. There simply was no testimony at trial to support these statements. In fact, Stolorow, Tahir, and the defense hair expert, Kenneth Siegesmund, testified that no such identification was possible.
The impropriety of such an argument was recognized in People v. Giangrande (1981), 101 Ill. App. 3d 397. There, as here, the State’s expert testified only that hairs found at the crime scene “could have originated from the defendant,” yet the prosecutor argued that defendant’s hairs had in fact been found there. (Giangrande, 101 Ill. App. 3d at 402-03.) The appellate court reversed the conviction, holding that the prosecutor had improperly “overstated the evidence.” (Giangrande, 101 *31Ill. App. 3d at 402-03.) We note that a pretrial colloquy in the record reveals that the State was aware of Giangrande, which makes the prosecutor’s improper comments here more objectionable.
The prosecutor’s misrepresentation of the hair-comparison evidence was compounded by his argument that the mathematical probabilities that the hairs found on the victim’s body and in her apartment came from anyone other than defendant were minuscule. The prosecutor relied on hair-comparison studies published by the forensic scientist Barry Gaudette for the statistics he used. The appellate court opinion concluded that “[plainly, the figures from Gaudette’s study do not have general application to hair comparisons” (159 Ill. App. 3d at 79) and that the prosecutor’s message to the jury, based on these odds, “was *** a distortion of [the witness’] testimony.” 159 Ill. App. 3d at 79.
The only testimony heard on the figures was elicited from Siegesmund on cross-examination. The entire text of the cross-examination follows:
“PROSECUTOR: You are aware of a forensic scientist by the name of Barry Gaudette, are you not?
SIEGESMUND: Gaudette is one of the proponents of X-ray analysis.
Q. Mr. Gaudette performed a study in the early to middle 70’s, did he not, with regard to the percentages and probabilities of hair comparisons?
A. Absolutely.
Q. And his technique that he used was with a comparison microscope, was it not sir?
A. He used comparison. And he, also, used other microscopes.
Q. But he used a comparison microscope. The one microscope you did not use, is that correct?
A. Yes, he did use that, also.
*32Q. And his probabilities came to the substance that a match between head hairs is likely in one out of every 4,500 cases, is that correct.
A. Well, can I explain that?
Q. I’ll rephrase the question. Did he not come up with a figure that any two individuals, the probability they would have matching head hairs is a likelihood in one out of 4,500?
A. It depends on how many hairs you are talking about.
Q. Would you say, the more hairs you have to compare, the closer to that figure you get?
A. The higher the probability, that is correct.
Q. So in this case if we had but one hair that Mr. Tahir linked to Mr. Linscott, that would have that much meaning, is that correct?
A. Yes. Using the conventional techniques that Gaudette used, yes.
Q. That Gaudette used, that’s correct?
A. Yes.
Q. If you have two to three hairs, your information is a little better, is that correct?
A. Yes. Only if you do the forty tests he recommends.
Q. Fine, if you had, approximately, seven or eight hairs, you have more information to base it on?
A. According to Gaudette, that would give you a higher probability. If you did the forty tests.” (Emphasis added.)
Based on the evidence at trial, the mathematical probabilities from Gaudette’s study should not have been considered by the jury. Siegesmund made it clear in his testimony that the findings in Gaudette’s studies were based on the completion of UO tests, not simply the 7 to 12 comparison tests performed by Tahir. (See Linscott, 114 Ill. 2d at 357 (Clark, C.J., concurring in part and dissenting in part).) There is no evidence that “forty tests” were performed in this case; therefore, there is no foundation for the assumption that Gaudette’s mathematical statistics applied.
*33In his closing argument, however, the prosecutor made the following statement to the jury:
“Then I asked [Siegesmund] if he was aware of the Chief Forensic Scientist for the Royal Canadian Mounted Police, Mr. Dejur? [sic (Gaudette)] And the man says yes, of course, he is the leading man in his field. And I asked him are you familiar with the figure formula and those were that of any two head hairs matched from two separate individuals it occurs in one out of every forty five hundred times.
I asked Mr. Siegesmund are you also familiar with the fact that one person may have the same head hair does not mean that he has the same pubic hair and that the pubic hair ratio in any two people matched are one in eight hundred and I leave it to you, Ladies and Gentlemen of the jury, to figure out the probability or .likelihood of anybody having the same head hair and the same pubic hair. It is a figure of one out of every forty-five hundred and one out of every eight hundred.”
The prosecutor’s comments were without foundation and were patently inapplicable to the results of the tests performed on the hairs in the case. Nonetheless, as was-argued by defendant in his brief to the court, the prosecutor all but told the jury that the odds of another individual’s having hair with the same characteristics as defendant’s hair were somewhere in the area of one in 3 million. In other words, the odds that another individual would have both head hair and pubic hair with the same characteristics as defendant’s head hair and pubic hair are the product of one in 4,500 multiplied by one in 800. In fact, not only was the jury told that the likelihood that anyone other than defendant had left hairs in the victim’s apartment was incredibly small, but the prosecutor told the jurors that the hairs found in the victim’s apartment, in the rug under her body, and in her pubic region were, conclusively, defendant’s hairs. We are compelled to find that these statements about the hair-com*34parison evidence, when considered by the jury which heard defendant testify that he had a dream on the night of the victim’s death paralleling the events of her death in some respects, so prejudiced defendant’s rights that they alone are reversible and require remanding the cause for a new trial.
The appellate court, however, concluded that other statements the prosecutor made in closing regarding the hair-comparison evidence were also improper. Specifically, the appellate court concluded that the prosecutor’s repeated claim that hairs found at the scene of the crime and on the victim’s body “matched” defendant’s hair was a “distortion of Tahir’s testimony, [and] also a distortion of the meaning of the evidence.” (159 111. App. 3d at 79.-) The opinion notes that these claims were especially misleading in light of the fact that Tahir declined to substitute the word “match” in place of “consistent” when the prosecutor attempted to do so:
“Q. So again, all we are talking about is that they match in every respect, is that correct?
A. They were consistent.”
159 Ill. App. 3d at 77.
In the prosecutor’s opening statement, he told the jury that “the hairs from Karen’s right hand, not one, not two, but more matched the defendant Steven Linscott.” The prosecutor also told the jury that “the two pubic hairs also matched the defendant Steven Linscott.” Explaining to the jury that defendant asked the police why he was being arrested, the prosecutor told the jury that “[defendant] was told because your hairs match.” In closing arguments, the prosecutor repeated this “matching” theme, arguing to the jury that “[p]ubic hair in the woman’s crotch match[ed] Mr. Linscottfs].” “The hairs,” the prosecutor argued, “tie him up to it.”
We also believe that the prosecutor’s use of the word “match” or “matched” misrepresented the scientific evi*35dence to the jury and was improper in light of the refusal of the State’s own expert to use the word when testifying about the evidence. Tahir’s testimony was simply that the hairs from defendant were “consistent” with those found on the victim and in her apartment. To tell the jury otherwise, that is, that the hairs matched, was not supported by the testimony at trial.
The prosecutor was entitled, of course, to summarize the testimony and urge the jury to convict defendant based on the evidence. However, to say that the hairs “matched” rather than were “consistent” was not based on the evidence. We are unpersuaded by the State’s claim that the words “matched” and “consistent” are “in some respects synonyms of each other.” While it may be true that all things that “match” are also “consistent,” it is not true that all things that are “consistent” also “match.”
Further, we believe use of the word “match” misrepresented the essential nature of hair comparison. Unlike a fingerprint, for example, an individual’s hair does not have characteristics that are unique only to his hair. In fact, as was revealed in Stolorow’s testimony, hairs on an individual’s head may vary from hair to hair. To say that the hair samples taken from defendant “matched” the hairs found on the victim’s body and in her apartment is to argue, essentially, that the source of these hairs was defendant. And, unlike a fingerprint, hair cannot be used to identify an individual.
Despite our analysis, however, we are satisfied that these “match” or “matched” remarks alone did not “ ‘ “constitute a material factor in the conviction.” ’ ” (Lyles, 106 Ill. 2d at 391, quoting Fields, 59 Ill. 2d at 522.) The prosecutor’s “match” comments were improper, but we do not find, as did the appellate court, that they were so “egregious” as to deny defendant a fair trial. 159 Ill. App. 3d at 74.
*36Likewise, we disagree with the appellate court’s conclusion that the prosecutor's use of the word “identical” was in error. In his closing argument, the prosecutor told the jury that Tahir had looked at “eighteen” different characteristics. He then argued:
“What does [Tahir] mean by consistent. He means there is nothing dissimilar.
I would suggest to you Ladies and Gentlemen if I said there were two American flags right there and they were both twelve by eight and that they had the same number of stars and stripes and had the same coloring you would sit there and say well there is nothing dissimilar about those two, they are identical.
But not a scientist. A scientist will state that every aspect that I examined they were consistent. And what does that mean. There was nothing different. And to a layman it means identical as the two American flags.”
The prosecutor did not use the word “identical” to overstate the evidence, or to improperly make it “more conclusive” than it was, as defendant would have this court believe. If the prosecutor can be faulted for overstating the hair-comparison evidence, it would not be with respect to his statements that the hairs compared “matched” or were “identical,” but rather, in telling the jury that the hairs found on the victim’s body and in her apartment were defendant’s hairs. We do not find the prosecutor’s use of the word “identical” (which appears only once in his statements) improper, because it appears in an analogy he created to explain the evidence to the jury. The prosecutor accurately summarized Tahir’s testimony about the meaning of “consistent,” and then, in argument, “suggested” to the jury what that testimony meant to them as laypersons. Having heard the testimony, the jury was free to reject the prosecutor’s analogy.
The appellate court also found the prosecutor made improper comments regarding the results of blood comparisons between defendant’s and the victim’s blood *37samples. (159 Ill. App. 3d at 75-76.) Blood-typing tests were performed on a sample of the victim’s blood and on the blood and saliva samples provided by defendant. As -to the victim, these tests revealed that she was a secretor and that her ABO blood type was 0. As to defendant, the tests revealed that he is a nonsecretor whose ABO blood type is AB.
“Secretors” and “nonsecretors” are classifications that refer to groups of individuals whose bodies do (secretors) or do not (nonsecretors) replicate their ABO blood types in body fluids other than blood. Because defendant is a nonsecretor, blood-typing tests performed on any of his body fluids other than blood, such as semen, would not reveal his ABO blood type. In circumstances where defendant’s body fluids were combined with the body fluids of another individual (as during sexual intercourse), blood-typing tests performed on the resulting mixture would reveal only the blood type of defendant’s partner.
Blood-typing tests were also performed on the vaginal swab taken from the victim. Tahir testified that the tests revealed the presence of seminal material and ABO blood type 0. The results of the tests performed on the vaginal swab supported the conclusion that the victim had sexual intercourse prior to her death with either a secretor with type 0 blood or a nonsecretor.
In opening and closing statements, however, the prosecutor simply ignored the fact that the seminal material could have come from a secretor with type 0 blood and repeatedly stated that it came from a nonsecretor. In the words of the appellate court opinion, the prosecutor “simply made up” the evidence that the seminal material on the vaginal swab came from a nonsecretor. (159 Ill. App. 3d at 76.) This “made-up evidence” was doubly devastating because not only was it false, but it reduced the pool of possible assailants to a relatively small percentage of *38the population, that is, nonsecretors. (See 159 Ill. App. 3d at 76.) Tahir testified that only 20% of the population are nonsecretors (contrary to the prosecutor’s statement that only 15% of the population fall into this category).
This contention first appears in the prosecutor’s opening statements:
“We expect the evidence to show that when an analysis on [defendant’s] saliva was done it turned out to be what they call a non-secretor. *** When the laboratory went further and tested the vaginal swab they couldn’t determine a blood type because it came back consistent with a non-secretor. Only fifteen per cent of the population.”
In closing argument, the prosecutor stated: “Karen was raped by a non-secretor and the defendant is a nonsecretor.” Later, he stated:
“[Defense counsel] says you must ask yourself where is the link. The link is the semen matching the non-secretor. [Defendant] is a non-secretor *** Seminal material. One fact it came from a non-secretor. [Defendant] is a non-secretor.” (Emphasis added.)
A prosecutor must confine his arguments to the evidence and to “reasonable inferences” that follow from it. (See, e.g., People v. Carlson (1982), 92 Ill. 2d 440, 449; People v. Beier (1963), 29 Ill. 2d 511, 517.) We believe that the prosecutor in the instant case contravened this fundamental rule. Evidence was presented that there were two possible sources of the seminal material — either a male secretor, with type 0 blood, or a male nonsecretor, with either type 0 or type AB blood. No evidence was presented, however, that the seminal material was more likely from one of these sources than from the other. In fact, Tahir testified that the results of the tests conducted on the swab supported both conclusions.
Therefore, the prosecutor’s conclusion that “Karen was raped by a nonsecretor” plainly cannot be inferred from the evidence. The only conclusion supported by the *39evidence is that the victim was raped by a nonsecretor or a secretor with type 0 blood. Also improper were the prosecutor’s arguments that “the vaginal swab *** came back consistent with a non-secretor” and that the “seminal material *** came from a non-secretor.” Although the evidence supports the proposition that a nonsecretor was a possible source of the seminal material found in the victim’s body, the existence of such a possibility does not permit the conclusion that the material could only have come from such a person.
We agree with the appellate court’s conclusion that the prosecutor misrepresented the blood-typing evidence in closing and find that these comments alone were sufficient to require reversal and remandment for a new trial. This court has previously adopted the reasoning of the appellate court decision on this issue. In People v. Turner (1989), 128 Ill. 2d 540, this court was asked to decide whether the prosecutor in that case had misrepresented the physical evidence connecting the defendant to the victim. In holding that he had not, this court referred to the Linscott opinion:
“In People v. Linscott (1987), 159 Ill. App. 3d 71, the prosecutor clearly mischaracterized the evidence when he told the jury that a vaginal swab indicated that the victim had been raped by a nonsecretor, a person whose blood type cannot be detected in his body fluids, and that the defendant was a nonsecretor. This evidence was made up by the prosecutor because no one testified that the victim was raped or that the seminal material found came from a nonsecretor.” (Turner, 128 Ill. 2d at 559.)
Although the court in Turner distinguished Linscott on its facts, this court did not reject the appellate court’s analysis.
In defense of the prosecutor’s statements regarding the blood-typing tests, both the State and the dissent below observe that a prosecutor may comment on the evi*40dence, even if contradictory evidence was presented. (159 Ill. App. 3d at 91 (McNamara, P.J., dissenting); People v. Gacy (1984), 103 Ill. 2d 1.) This is true but irrelevant. Evidence was presented at trial that there were two alternative sources of the seminal material. Alternative evidence is certainly not to be construed as contradictory evidence, allowing a prosecutor to improperly argue unsupported conclusions to a jury. The question here is whether the prosecutor’s argument reasonably followed from the testimony of the State’s own experts. We conclude that the answer is no.
Finally, the dissent below concludes that the “brief comments” made by the prosecutor with regard to the blood-typing test results do not constitute reversible error in light of the length of the trial. (159 Ill. App. 3d at 91 (McNamara, P.J., dissenting).) Moreover, the dissent concludes that “[t]he majority dramatizes and magnifies the effect of the prosecutor’s language by placing the two quotations next to each other when they are actually separated by 63 pages ***.” 159 Ill. App. 3d at 91 (McNamara, P.J., dissenting).
There were in fact four (not two) prejudicial comments regarding the blood-typing test results: one in the opening statement, one in the opening portion of the closing argument, and two in rebuttal. However, we believe that more significant than the number of prejudicial comments is the fact that every one of the prosecutor’s comments regarding the blood-typing evidence misrepresented the evidence. The weight of the evidence in a case is a relevant consideration in determining whether the remarks prejudiced the defendant. The closer the evidence is in the case, the greater the likelihood of prejudice. (See, e.g., Lyles, 106 Ill. 2d at 391; People v. Weaver (1982), 92 Ill. 2d 545, 560.) We believe that the evidence in this case was closely balanced, making the *41prejudicial impact of the prosecutor’s misrepresentations of the blood-typing evidence more significant.
Finally, in remanding for a new trial on the murder charges, the appellate court, sua sponte, raised the issue of the relevancy of the vaginal swab evidence. The court held that “the vaginal swab evidence would not be relevant as substantive evidence in the new trial of the murder charge.” (159 Ill. App. 3d at 84.) Even if the swab were to be considered relevant, the majority suggested that it would be subject to exclusion because of “the State’s advertent destruction of the vaginal swab.” (159 Ill. App. 3d at 84 n.7.) Alternatively, the court held that “to the extent that the vaginal swab evidence could possibly have any probative value, its probative value is substantially outweighed by the danger of unfair prejudice that would result from the admissibility of that evidence.” 159 Ill. App. 3d at 85.
Although the substantive value of the swab may be limited, this alone does not require that the evidence be excluded at a new trial on the murder charge. The State argues that, although it is of limited value, the swab evidence is relevant because it has a “tendency to prove that defendant was a personal acquaintance of the victim and therefore could have been her murderer.” We agree. Granted, the results performed on the swab were inconclusive as to the source of the seminal material. However, the results failed to exclude defendant from the pool of males who may have "had a sexual encounter with the victim. Consequently, the results also failed to eliminate defendant as a suspect in the victim’s homicide.
Because the appellate court considered the relevance of the swab evidence sua sponte, neither party before this court has had the opportunity to adequately address the issues of either the evidence’s relevance on the murder charge alone or the relative prejudice that would result in the event the evidence were excluded from a *42new trial. We do not believe the appellate court is the proper forum to first consider these issues, especially when the review is initiated by that court, without benefit of briefing or argument by the parties. These issues should be considered by the trial court on remand before they are considered by a reviewing court, if a review of the trial court’s decision on these issues is ever in fact sought. We conclude that exclusion of the evidence is a matter to be decided by the trial court on remand.
For the foregoing reasons, the judgment of the circuit court is reversed and the cause is remanded to that court for a new trial consistent with this opinion. Because the appellate court reversed the circuit court judgment but included directions in the judgment line of its opinion, that judgment is vacated.
Appellate court judgment vacated; circuit court judgment reversed; cause remanded.
JUSTICES BILANDIC, HEIPLE and FREEMAN took no part in the consideration or decision of this case.