OPINION OF THE COURT

The question before this Court is whether the Superior Court erred in determining that the Commonwealth presented sufficient evidence to support Appellant’s conviction of driving under the influence of alcohol pursuant to 75 Pa.C.S. § 3731(a)(4) (operation of a motor vehicle while the amount of alcohol by weight in the blood of the person is 0.10% or greater). Upon review of the record we conclude that the evidence was insufficient to establish that Appellant possessed *396a blood alcohol level of .10% or greater at the time he operated a motor vehicle. Therefore, we reverse the decision of the Superior Court.

In reviewing a challenge to the sufficiency of the evidence we must determine whether, viewing all the evidence admitted at trial, together with all reasonable inferences to be drawn therefrom, in the light most favorable to the Commonwealth as verdict winner, the jury could have found that each element of the charged offense was proved beyond a reasonable doubt. Commonwealth v. Hughes, 521 Pa. 423, 555 A.2d 1264 (1989).

On September 28, 1990 a police officer was engaged in an unrelated traffic stop when a portion of Appellant’s motor vehicle struck the police officer as Appellant passed the police officer. The police officer radioed for assistance, commenced pursuit, and subsequently stopped Appellant. At this time Appellant’s speech was slurred, his eyes were bloodshot and watery, and he emitted an odor of alcohol. Appellant refused field sobriety tests and demanded a blood test. He was immediately arrested. A blood alcohol content (“BAC”) test performed at a local hospital approximately two hours later provided a BAC of .141%. Appellant was subsequently charged with reckless driving,1 disorderly conduct,2 and two counts of driving under the influence of alcohol.3 The charge of disorderly conduct and one count of driving under the influence of alcohol (75 Pa.C.S. § 3731(a)(1)) were subsequently dismissed at Appellant’s preliminary hearing. A jury trial commenced on November 13, 1991, and Appellant was found guilty of driving under the influence of alcohol pursuant to 75 Pa.C.S. § 3731(a)(4).

On appeal to the Superior Court, Appellant argued that the evidence presented was insufficient to support his conviction because it failed to establish that he had operated his vehicle while possessing a BAC of .10% or greater in accordance with this Court’s decisions in Commonwealth v. Jarman, 529 Pa. *39792, 601 A.2d 1229 (1992) and Commonwealth v. Modaffare, 529 Pa. 101, 601 A.2d 1233 (1992).

In Jarman and Modaffare, this Court reversed the judgments of sentence arising out of drunk driving convictions pursuant to 75 Pa.C.S. § 3731(a)(4), because the BAC evidence introduced by the Commonwealth did not sufficiently establish that the BAC of the defendants exceeded permissible limits at the time the defendants were driving. We reached this conclusion because the offense defined by Section 3731(a)(4) is to drive a vehicle while one’s blood alcohol content is .10% or greater, and delay in BAC testing in those cases could have reflected BAC results that were significantly higher than those possessed while the defendants were operating their motor vehicles. In other words, the defendants’ BAC could have been below the .10% threshold at the time they were stopped and could have simply risen above the .10% threshold as a result of alcohol being absorbed into the defendants’ blood during the delay prior to BAC testing.4

Based upon the lapse of time between driving and BAC testing; the BAC test results that narrowly exceeded the .10% threshold; a ten percent margin of error associated with BAC testing; and the expert testimony provided by the Commonwealth that the BAC of the defendants could have been below .10% while they were driving, we concluded that the jury was permitted to engage in unbridled speculation as to whether the BAC of the defendants were at or above .10% at the critical time of operation. The majority thereafter went on to recognize a strong inference that an accused’s BAC would be in the prohibited range while driving where he possesses a BAC significantly above .10% and where BAC testing oc*398curred soon after driving terminates. However, it did not specifically define what blood alcohol level is significantly above .10% or a temporal cut off for the lapse of time between driving and BAC testing.

In the matter sub judice, the Superior Court correctly interpreted our decisions in Jarman and Modaffare through its reliance on Commonwealth v. Osborne, 414 Pa.Super. 124, 606 A.2d 529 (1992), allocatur denied, 531 Pa. 660, 613 A.2d 1209 (1992), which provides:

In our view, and in accordance with the dual standards set by our supreme court in Jarman and Modaffare, the stronger the inference of guilt, the less significant is the necessity for evidence of relating back. Conversely, the weaker the inference of guilt, the more vital is the necessity for evidence of relating back an accused’s BAC test result to the time of driving.

Id. at 128, 606 A.2d at 531.

Based upon application of these factors, the Superior Court herein concluded that the trial court had erred in failing to properly apply Jarman and Modaffare, and that Appellant’s BAC test results of .141% obtained after a two hour delay did not give rise to a sufficiently strong inference that Appellant possessed a BAC of .10% at the time he drove. We agree with the Superior Court that because Appellant’s BAC did not represent a substantial departure from the permissible limit, and his BAC may have been below .10% at the time he was stopped and risen above this limit during the substantial delay prior to testing, the inference of guilt was too weak to support Appellant’s conviction in the absence of evidence relating his BAC test results back to the time of driving.

Nevertheless, the Superior Court then went on to determine that there existed additional evidence from which the jury could infer that Appellant’s BAC was in excess of .10% at the time when he was stopped, and that Appellant’s BAC did not rise between the time he was stopped and the time the BAC test was performed. In other words, the Superior Court determined that other evidence existed which would fulfill the *399“relation back” requirements of Jarman and Modaffare. This evidence consisted of the testimony of two police officers during the Commonwealth’s case in chief that upon being stopped Appellant slurred his speech, which served as the basis for an opinion of Appellant’s own expert on cross-examination that most people do not begin to exhibit visible signs of intoxication until they achieve a BAC of approximately .15%.5 On this basis, the Superior Court determined that the jury’s finding that Appellant possessed a BAC of .10% or greater while operating his motor vehicle was not based solely on “unbridled speculation” as in Jarman and Modaffare.

Before us, Appellant contends that the Superior Court erred in determining that the existence of this additional evidence sufficiently related Appellant’s BAC test results back to the time when he was driving, because such evidence was irrelevant, and thus, could not serve to remedy BAC test results that were found to be insufficient under Jarman and Modaffare. We agree.

*400The offense of driving under the influence of alcohol or controlled substance is set forth at 75 Pa.C.S. § 3731.6 Based upon this provision, in charging Appellant with the unlawful operation of a motor vehicle while under the influence of alcohol, the Commonwealth was authorized by the Legislature to proceed on either or both of two theories. It could allege that Appellant was under the influence of alcohol to a degree which rendered him incapable of safe driving pursuant to subsection (a)(1). It also could allege that the amount of alcohol by weight in Appellant’s blood was .10% or greater pursuant to subsection (a)(4). In fact, Appellant was initially charged with violating both of these provisions. However, at Appellant’s preliminary hearing the charge alleging a violation of subsection (a)(1) was dismissed. Therefore, the Commonwealth proceeded to trial only on the charge alleging a violation of subsection (a)(4), which limited the inquiry to whether Appellant operated a motor vehicle while the amount of alcohol by weight in his blood was .10% or greater. 75 Pa.C.S. § 3731(a)(4).

Appellant contends that the manner in which the Commonwealth may prove a violation of subsection (a)(4) is limited by the nature of the inquiry to scientific BAC testing evidence, since it is the only manner in which a BAC of .10% or greater may be established. Thus, if scientific BAC testing is the only evidence that may used to support a conviction pursuant to subsection (a)(4), then where such evidence is insufficient, a *401conviction pursuant to subsection (a)(4) cannot be upheld based upon additional evidence that an accused manifested symptoms of intoxication. In support of this proposition Appellant cites to the decision of the Superior Court in Commonwealth v. Kemble, 413 Pa.Super. 521, 605 A.2d 1240 (1992), allocatur denied, 532 Pa. 651, 615 A.2d 340 (1992).

In Kemble the question presented was whether the trial court erred in admitting, and/or in instructing the jury with regard to, evidence provided by a police officer in the Commonwealth’s case in chief concerning the defendant’s impaired physical condition. Specifically, a police officer testified that he observed the defendant make an extremely wide turn, nearly proceed down a one-way street, and veer directly in front of his patrol car. The officer then testified that the defendant had difficulty determining which document was her registration card, staggered when she exited her automobile, possessed an odor of alcohol, leaned against a building and failed two sobriety tests. Finally, the officer testified that subsequent to an intoxilyzer test the defendant vacillated between an “emotional” mood and a “combative” mood, and was uncooperative.

On appeal from the defendant’s conviction, the Superior Court determined that the trial court abused its discretion in admitting this “impairment” evidence, and awarded the defendant a new trial. The Superior Court concluded that such evidence was irrelevant to the inquiry involved under subsection (a)(4), and that such evidence could have been particularly prejudicial to the defendant, since the defendant had disputed the validity of the BAC test results by claiming that she drank water immediately prior to performing the BAC test. In addressing the question of the relevancy of the “impairment” evidence, the Superior Court stated:

Therefore, any evidence that is relevant to prove impairment is admissible in a § 3731(a)(1) prosecution. However, a § 3731(a)(4) charge, when unaccompanied by a § 3731(a)(1) charge, involves a narrowly focused inquiry that is solely concerned with a determination of whether a defendant had a .10% or greater blood alcohol level at the *402time that she drove her automobile. During such an inquiry impairment evidence is not relevant since it does not logically or reasonably tend to prove or disprove that a defendant’s blood alcohol level was .10% or greater at the time that she drove her automobile, it does not tend to make such a fact more or less probable and it does not afford a basis for or support a reasonable inference or presumption regarding whether a defendant’s blood alcohol level was .10% or greater. As mentioned, supra, the admission of the impairment evidence could have been particularly prejudicial to appellant since the validity of the results of her intoxilyzer test were disputed at trial. Since the impairment evidence was not relevant the trial court abused its discretion in admitting said evidence. Hence, we remand this case for a new trial.

Id. at 526, 605 A.2d at 1242. (citations omitted).

We agree with the decision in Kemble, and find no basis upon which to conclude that evidence beyond that provided by scientific testing may in any manner satisfy the inquiry raised by a charge pursuant to subsection (a)(4) where the Commonwealth is not also advancing a charge pursuant to subsection (a)(1). Because such evidence is not scientific evidence, it cannot make it more probable that an accused possessed a BAC of .10% or greater at the time he operated his motor vehicle where BAC test results are alone insufficient. Rather, it can serve only to prejudice an accused’s right to have the prosecution satisfy its burden of proving that he possessed a blood alcohol content of .10% or greater at the time he operated a motor vehicle. Quite simply, given the framework of Section 3731, evidence beyond that provided by scientific BAC testing is not relevant to a determination of whether an accused violated subsection (a)(4).

To hold otherwise would eliminate the distinction between subsection (a)(4) and subsection (a)(1), which by virtue of the enactment of subsection (a)(4) the Legislature clearly could not have intended. Subsection (a)(1) is a general provision and provides no specific restraint upon the Common*403wealth in the manner in which it may prove that an accused operated a vehicle under the influence of alcohol to a degree which rendered him incapable of safe driving. However, in 1982, when the Legislature amended Section 3731 by adding subsection (a)(4), it defined one form of evidence, BAC test results of .10% or greater, which would conclusively establish an element of the offense of driving under the influence of alcohol. That is, an accused is under the influence of alcohol to a degree that renders him incapable of safe driving as a matter of law if his BAC is .10% or greater. This definition, by its very nature and existence requires that the Commonwealth satisfy its burden solely through competent BAC test results in order to support a conviction pursuant to subsection (a)(4). In the absence of such evidence, the prosecution, by necessity, becomes one pursuant to subsection (a)(1). If it does not, subsection (a)(4) would be rendered meaningless in violation of the well established principle of statutory construction that every statute shall be construed, if possible, to give effect to all its provisions. 1 Pa.C.S. § 1921(a).

Accordingly, we hold that where, as in the matter sub judice, an accused is not defending a charge of violating subsection (a)(1) evidence beyond scientific BAC testing is not only irrelevant, but prejudicial. Appellant’s BAC test results alone were insufficient to establish a violation of subsection (a)(4), and no abundance of impairment evidence could have remedied this shortcoming. Rather, such could serve only to have prejudiced Appellant by swaying a jury, already forced into speculation by the insufficiency of the BAC test results, to find that Appellant possessed a BAC of .10% or greater at the time he operated a motor vehicle in violation of subsection (a)(4).7

*404The decision of the Superior Court is reversed, and the judgment of sentence imposed by the Court of Common Pleas of Berks County, Criminal Division, is vacated.

PAPADAKOS, J., did not participate in the decision of this case.

ZAPPALA, J., concurs in the result.

CASTILLE, J., files a dissenting opinion.

MONTEMURO, J., is sitting by designation.

dissenting.

The majority holds that circumstantial evidence is not relevant to establish blood alcohol content (BAC) in a prosecution for driving under the influence of alcohol in violation of 75 Pa.C.S. § 3731(a)(4). Because I believe that neither the statute nor the case law prohibits the use of circumstantial evidence in relating back appellant’s BAC to the time of driving, I must respectfully dissent.

It is well established that the Commonwealth can prove any or all elements of an offense through circumstantial evidence. Commonwealth v. Zimmick, 539 Pa. 548, 555, n. 9, 653 A.2d 1217, 1220-21, n. 9 (1995) (citing Commonwealth v. Hardcastle, 519 Pa. 236, 250, 546 A.2d 1101, 1107-08 (1988)). Under 75 Pa.C.S. § 1547(c), BAC can be established through the *405results of chemical testing. However, nothing in either that section or in 75 Pa.C.S. § 3731(a)(4) restricts the proof of BAC to chemical tests. To the contrary, the legislature specifically provided further that subsection (c) “shall not be construed as limiting the introduction of any other competent evidence bearing upon the question of whether or not the defendant was driving under the influence of alcohol.” 75 Pa.C.S. § 1547(f). By holding that circumstantial evidence cannot be used to establish BAC related back to the time of driving, the majority nullifies this section of the statute.

Furthermore, neither Commonwealth v. Jarman, 529 Pa. 92, 601 A.2d 1229 (1992), nor Commonwealth v. Modaffare, 529 Pa. 101, 601 A.2d 1233 (1992) prohibit the use of circumstantial evidence to prove BAC. Rather, the Court held that, where the BAC barely exceeded .10% and there had been a significant time lapse between driving and the blood test, the inference of guilt was weakened. Jarman, supra, 529 Pa. at 96, 601 A.2d at 1231; Modaffare, supra, 529 Pa. at 105, 601 A.2d at 1235. Under the circumstances of those particular cases, the Court held that the evidence was insufficient to establish relation back of the appellants’ BAC to the time of driving. Jarman, supra, 529 Pa. at 97, 601 A.2d at 1231; Modaffare, supra 529 Pa. at 106-07, 601 A.2d at 1236.

In Jarman, where the appellant’s BAC was .114%, just barely over the legal limit, one hour after driving, an expert testified that if appellant had consumed his last drink at the time he testified, his BAC would still have been rising at the time of testing, and would probably have been below .10% at the time of driving. In Modaffare, where appellant’s BAC was .108% one hour and fifty minutes after driving, the Commonwealth’s expert testified that there was no evidence upon which he could form any opinion as to what the alcohol level was at the time of appellant’s accident. The Court held in both cases that given the delay in testing appellants’ BAC, the slight deviation from the legal limit, and the lack of evidence from which the expert could form an opinion that appellant’s BAC at the time of the accident had exceeded .10%, there was not sufficient evidence to sustain the convic*406tions. Jarman, supra, 529 Pa. at 97, 601 A.2d at 1231; Modaffare, supra, 529 Pa. at 106-07, 601 A.2d at 1236.

The present case is distinguishable from both Jarman and Modaffare. Here, appellant’s BAC was .141%, more than 40% above the legal limit.1 Furthermore, appellant’s own expert was able to form an opinion, based upon the evidence presented at trial, that appellant’s BAC was in fact greater than .10%, and probably .15% or greater, at the time of driving. This appears to be precisely the sort of evidence, missing in Jarman and Modaffare, which would have rehabilitated the inference of guilt weakened by a borderline BAC and the delay in testing.

By excluding the use of circumstantial evidence in relating back BAC to the time of driving, the majority is requiring that the Commonwealth prove its case not just beyond a reasonable doubt, but by a mathematical certainty. Given that alcohol absorption rates differ with the individual, the result of this holding is that, as Mr. Justice Cappy stated in his dissent to Jarman,

Every time there is a case in which the test results are between .10 and .15 percent (approximately), and the test was not performed within minutes of the actual driving, the Commonwealth will be unable to prove its case. This would defeat the vital purpose and intent of the statute and would operate as a de facto evisceration of the .10 percent blood alcohol limitation in the statute.

529 Pa. at 100, 601 A.2d at 1233 (Cappy, J., dissenting).

Accordingly, I would affirm the order of the Superior Court, affirming the judgment of sentence.