Opinion

Kenneth Palestine Davis was convicted in a bench trial pursuant to Code § 18.2-266(i) of driving a motor vehicle while he had a blood alcohol concentration of .10 percent or more by weight by volume as indicated by a chemical test administered in accordance with the provisions of Code § 18.2-268. On appeal, Davis phrases the issues to be addressed in the following manner: (1) whether, in accordance with the language of Code § 18.2-266(i), the Commonwealth met its burden of proof in establishing that Davis’ blood alcohol content was .10 percent or more while he was operating his vehicle; and (2) whether the trial court correctly ruled that Code § 18.2-266(i) precludes an ac*293cused from introducing evidence of his condition at the time of the alleged incident.

The essential facts leading to the charge against Davis are not in dispute. On November 24, 1986, at some time prior to 9:00 p.m. on Route 29 in Madison Heights, a truck driven by Davis struck the rear of a truck driven by John Hines, which was stopped at a stop light. Trooper Lynn Roach, a Virginia State Police officer, arrived at the scene of the accident at 9:15 p.m. and interviewed Davis and Hines. Hines related to Trooper Roach that Davis had attributed the accident to faulty brakes on his truck. Hines further related that he did not detect an odor of alcohol about Davis and that Davis was polite and apologetic about the accident. Davis, however, related to Trooper Roach that “he had had one beer in Danville, Virginia, as he was coming up on his trip from Danville up 29 through Lynchburg going on North up 29.” Davis stated that he had nothing alcoholic to drink after the accident. After administering a field sobriety test, Trooper Roach arrested Davis at 9:45 p.m. Subsequently, a blood test was administered on Davis at 10:17 that evening. The test results were reported as .10 percent by weight by volume.

At trial, Davis conceded that he was driving at the time of the accident and that the blood test was administered “in accordance with the provisions of Code § 18.2-268.,1 He also concedes these facts on appeal. However, because the test results were reported precisely at .10 percent, the minimum proscribed by Code § 18.2-266(i), he challenges the accuracy of these results and asserts that they did not accurately reflect his blood alcohol concentration at the time he was driving his truck.

In this context, it is apparent from the record that Davis intended a three-pronged defense to the charge. First, while conceding that the blood test was properly administered, he intended to challenge the precision of the test results based on standard devia*294tions inherent in the test procedures. Second, he intended to introduce evidence that the test results only established his blood alcohol concentration at the time the test was performed and not necessarily at the time he was driving. Third, he intended to introduce evidence that at the time the accident occurred, regardless of the subsequent blood test results, he was not under the influence of alcohol and rather the accident resulted from faulty brakes on his truck.

In anticipation of his defenses, Davis filed a pretrial motion to determine whether evidence of his “condition during the accident” would be admissible at trial. By order entered on June 30, 1987, the trial court ruled that Davis would not be permitted to introduce evidence relating to his “condition at the time of the alleged offense” because his condition at that time “is not at issue in this cause.” The trial court’s order further provided, however, that the “admissibility of evidence concerning the accuracy and credibility of the blood alcohol content [test] of the defendant at the time of the alleged offense” would be ruled upon at trial. At trial, the court reiterated its prior ruling and attempted to clarify it by indicating that, because Davis was charged under Code § 18.2-266(i) and not with driving under the influence of alcohol under Code § 18.2-266(ii), the evidence of his “condition” at the time of the accident was not “relevant.” The court also ruled that, because Davis was admittedly the driver of the truck, under Code § 18.2-266 (i) the issues at trial were limited to whether the blood test was administered in accordance with the provisions of Code § 18.2-268 and whether Davis “had 0.10 or not.”

It is apparent from the record that Davis’ imprecise references to his “condition during the accident” or “at the time of the alleged offense” made it difficult for the trial court to respond to his motion. Our review of the record, however, convinces us that the trial court ruled that a conviction under Code § 18.2-266(i) requires only that the Commonwealth prove beyond a reasonable doubt that Davis was driving the truck, and the blood test subsequently administered in accordance with the provisions of Code § 18.2-268 accurately reflected a blood alcohol concentration of at least .10 percent at the time the test was performed. Thus, consistent with the Commonwealth’s position at trial and on appeal, the trial court adopted what has come to be known as the “per se” application of Code § 18.2-266(i) and similar statutes in our sis*295ter states. That is, the trial court limited the issues to the accuracy of the test at the time it was performed rather than the accuracy of the test as reflecting blood alcohol concentration at the time the accused was driving. In addition, pursuant to this construction of the statute, the trial court did not consider Davis’ general physical condition, that is, whether he was driving while under the influence of alcohol, regardless of the blood test results.

To determine whether the trial court correctly construed the statute so as to limit the issues at trial, we must analyze the statutory scheme embodied in Code § 18.2-266 and related Code sections. While our Supreme Court has not had the occasion to address the issues presented in this appeal under Code § 18.2-266(i), a brief review of the legislative history of these Code sections reveals their legislative purpose.

At the time of the proceedings below, Code § 18.2-266 provided in pertinent part:

It shall be unlawful for any person to drive or operate any motor vehicle . . . (i) while such person has a blood alcohol concentration of 0.10 percent or more by weight by volume as indicated by a chemical test administered in accordance with the provisions of § 18.2-268, or (ii) while such person is under the influence of alcohol ....

(emphasis added).

Prior to 1984, Code § 18.2-266 prohibited driving “while under the influence of alcohol” (or other drugs or intoxicants). In order to facilitate that determination, when chemical tests were performed, Code § 18.2-269 provided for certain presumptions for and against the conclusion that a driver was “under the influence of alcohol” at the time of the alleged offense. Specifically, subsection (3) provided that a test result of .10 percent or higher created a presumption that a driver was under the influence of alcohol. This presumption, however, was not conclusive and could be rebutted by competent evidence. See Sinault v. Commonwealth, 228 Va. 269, 271, 321 S.E.2d 652, 654 (1984).

It is a matter of common knowledge based on human experience that outward manifestations of intoxication will vary from individual to individual. While one highly intoxicated individual *296may exhibit few, if any, outward manifestations of intoxication, another individual may appear to be very intoxicated after consuming a small quantity of alcohol. Thus, a determination whether a person was “under the influence of alcohol,” even when aided by statutory presumptions, cannot be reduced to a readily usable mathematical or objective formula. Rather, that determination must be based upon the totality of the evidence. Under such circumstances the difficulty of establishing proof beyond a reasonable doubt that a person was under the influence of alcohol, and conversely, the difficulty of defending against an unwarranted charge, are readily apparent.

Undoubtedly, in part because of these difficulties but primarily in response to the public safety concerns over driving while intoxicated, in the 1980s state legislatures enacted legislation to strengthen the current laws against driving while under the influence of alcohol. Central to that legislation was the use of chemical tests to prescribe a maximum blood alcohol level beyond which driving would be unlawful. In theory, such tests eliminated the difficulties inherent in a subjective determination whether a person was “under the influence of alcohol.” The use of a chemical test is appealing because of its objectivity and certainty. Scientific advancements in the understanding of blood alcohol concentration coupled with the language employed in specific legislation have, however, complicated legislative attempts to deal with drinking and driving and resulted in various interpretations given to similar legislation in our sister states. Based on scientific evidence, courts have come to accept as a matter of common knowledge that blood alcohol concentration, as measured by a chemical test, is a function of many factors including, but not limited to, the amount of alcohol consumed, the amount and type of food in the stomach, the body’s alcohol absorption rate, an individual’s size, weight, age, stomach and liver condition, and the length of time between drinking and measurement. See, e.g., State v. Murphy, 453 N.E.2d 1304 (Ohio Mun. 1983); State v. Cooke, 270 N.C. 644, 155 S.E.2d 165 (1967). Courts also have come to accept that a given blood alcohol concentration will not always reflect the extent to which an individual’s ability to drive safely has been impaired. In short, a blood alcohol concentration which seriously impairs one individual’s ability to drive safely may impair to a different degree another’s ability to do so. Nevertheless,

*297[i]t has been demonstrated empirically, and it is now widely accepted, that a motorist’s ability to drive safely is adversely affected by a blood-alcohol content of .10%, even though some individuals may exhibit few outwardly visible symptoms of intoxication at that level .... Consequently, many states have legislated directly against driving with a blood-alcohol content of .10%, some defining it as a separate offense and some treating it as driving under the influence per se.

State v. Knoll, 718 P.2d 589, 591-92 (Idaho App. 1986)(citations omitted); see also State v. Ulrich, 478 N.E.2d 812 (Ohio App. 1984). Regardless of these distinctions, the tragic consequences to our society occasioned by drinking and driving are well known. Recognizing this, and consistent with the national trend, our legislature, beginning in 1984, made significant revisions to Code § 18.2-266 and related code sections.

In 1984 the legislature revised Code § 18.2-266 into subsections and enacted subsection (i), prohibiting driving while having a blood alcohol concentration of . 15 percent or more by weight by volume as indicated by a chemical test administered in accordance with the provisions of Code § 18.2-268. In 1986 this minimum blood alcohol concentration was lowered to the present .10 percent. In addition, the presumptions from the blood alcohol concentration as indicated by chemical tests administered pursuant to Code § 18.2-269 were made applicable only to Code § 18.2-266 (ii), the driving “under the influence” subsection of that Code section. Finally in 1987, effective April 1, 1988, Code § 18.2-266 was amended to provide that the degree of intoxication prohibited is “to a degree which impairs [the driver’s] ability to drive or operator any motor vehicle . . . safely.”

The legislative purpose of these amendments was to provide a statutory scheme within the provisions of Code § 18.2-266 and related code sections to prohibit drinking and driving under either of two separate and distinct circumstances. Code § 18.2-266 (ii) prohibits driving “while under the influence of alcohol.” In order to facilitate that determination, when chemical tests are performed, Code § 18.2-269 continues to provide for certain presumptions for and against the conclusion that a driver was “under the influence of alcohol” at the time of the alleged offense, that is, *298when he was driving.

In contrast, Code § 18.2-266(i) prohibits driving while the driver has a blood alcohol concentration of .10 percent or more as measured by a chemical test administered pursuant to Code § 18.2-268. The presumptions contained in Code § 18.2-269 are not applicable to this offense as a result of the 1986 amendment to this code section. Thus, the inquiry under Code § 18.2-266(i) is not whether a driver was in fact “under the influence of alcohol” to a degree that his ability to drive safely was affected; rather, the issue is whether at the time he was driving his blood alcohol concentration was at least .10 percent as measured by a subsequently administered chemical test. It is for this reason that subsection (i) and similar statutes in our sister states have come to be known as “per se” statutes. The use of objective chemical tests lends credence to such references. However, while the use of the short-hand reference, “per se,” is a convenient means for distinguishing Code § 18.2-266(i) from subsection (ii), this term can be misleading if used as descriptive of the offense embodied in subsection (i). Specifically, the term suggests that a blood alcohol measurement prescribed by subsection (i) is conclusive proof of the offense. For the reasons that follow, we hold that it is not; rather, we hold that the prescribed measurement is an evidentiary fact which creates a rebuttable presumption that the measurement accurately reflects the blood alcohol concentration at the time of driving.2

Because the evil which Code § 18.2-266(i) is intended to prohibit is driving with a specified blood alcohol concentration and because the language employed in the statute refers to driving while having a specified blood alcohol concentration, the after-administered blood alcohol concentration test results must be related to the consumption of alcohol before or during the act of driving. Admittedly, the literal terms of the statute are susceptible to an interpretation that would require the Commonwealth, after obtaining an accurate, blood test result of .10 percent or more, to produce additional scientific evidence or expert testimony that correlates this test result to the blood alcohol concentration at the time of driving. Such a result was reached in State v. Rollins, 141 *299Vt. 105, 444 A.2d 884 (Vt. 1982). Other courts have determined that the test results reflect a legislative determination that tests performed within specific time limitations after driving accurately reflect blood alcohol levels at the time of driving; that the test results establish a prima facie case of guilt; that the test results are conclusive; or that the tests are not conclusive and the accused may offer evidence opposing the tests. See, e.g., State v. Ulrich, 478 N.E.2d 812 (Ohio App. 1984); Mosley v. State, 365 S.E.2d 451 (Ga. App. 1988); State v. Tishio, 506 A.2d 14 (N.J. Super. A.D. 1986); Washington v. District of Columbia, 538 A.2d 1151 (D.C. App. 1988). The analysis by the courts in these cases and the discussion of the “per se” concept are instructive, but only in the context of consideration of the specific language employed by the legislative enactments in these states. Thus, to determine the proper interpretation of Code § 18.2-266(i), we must focus on the specific language used in that Code section, and the appropriate evidentiary weight of the results of the subsequently administered blood alcohol test.

Code § 18.2-266(i), by its literal terms, does not prohibit driving after consuming sufficient alcohol to register .10 percent or more by weight by volume as indicated by a subsequently administered chemical test within a specified time in accordance with the provisions of Code § 18.2-268. See State v. Rose, 323 S.E.2d 339 (N.C. 1984). The enactment of such a statute, clearly within the province of the legislature, would be accurately a “per se” offense. Guilt would be established by proof that the accused was driving or operating and that the results of the subsequently administered chemical test were .10 percent or higher. The relationship between the test results and the blood alcohol level at the time of driving or operating would be immaterial; no presumptions would be involved. The blood alcohol concentration at the time of the chemical test, which can be accurately determined at that time, would be conclusive, or “per se,” evidence of the prohibited alcohol concentration. Code § 18.2-266(i), however, was not written in this fashion. Nevertheless, this code section must be interpreted in the context of the factual circumstances in which, by its very terms, it is applicable. While the statute proscribes no time limit within which the chemical test must be administered after driving, it is axiomatic that there can be no prosecution under this statute without the existence of a chemical test obtained under the provisions of Code § 18.2-268(B). In addition, *300where there is evidence that alcohol has been consumed after driving the chemical test cannot accurately reflect the blood alcohol concentration at the time of driving. The chemical test simply cannot distinguish between two sources of alcohol. Conversely, where no alcohol is consumed between the time of driving and the time the chemical test is administered the test results can reflect only that alcohol consumed before or during driving. Code § 18.2-266(i) is applicable only in these latter circumstances. Thus consistent with the factual circumstances in which this Code section is applicable, the “as indicated by a chemical test” (emphasis added) language in this Code section provides the basis for a presumption that the blood alcohol concentration while driving was the same as indicated by the results of the subsequent test. However, because the accepted knowledge of blood alcohol absorption and elimination rates makes it clear that the timing of the chemical test is critical in terms of the accuracy of the test to reflect the prior blood alcohol concentration at the time of driving, the test results cannot be conclusive. Rather, because the blood alcohol concentration reflected by the chemical test necessarily resulted from alcohol consumed prior to or during driving, the test results are presumptive evidence of the blood alcohol concentration at the time of driving. As such, the accused may challenge the test results by competent evidence, such as, for example, that he had not consumed enough alcohol in the relevant time to reach the level indicated by the chemical test results. See Washington v. District of Columbia, 538 A.2d 1151 (D.C. App. 1988). Unless rebutted, however, the test results are sufficient to establish the blood alcohol concentration at the time of driving.

With these principles to guide us, we review the issues presented by Davis on appeal. To the extent that Davis offered evidence or attempted to offer evidence that faulty brakes caused the accident and that his “condition” was such that he was not “under the influence of alcohol” at the time he was driving his truck, the trial court correctly ruled such evidence was not relevant to the charge under Code § 18.2-266(i). This evidence was not relevant because it did not tend to prove that the blood test results did not accurately reflect his blood alcohol concentration at the time he was driving. Moreover, whether Davis was “under the influence of alcohol” was not at issue.

*301Davis was permitted to offer evidence from Dr. James C. Valentour, a toxicologist with the Central Office of the Bureau of Forensic Science. Dr. Valentour had eleven years of experience with the equipment used by that agency to make blood alcohol tests and was stipulated by the Commonwealth to be an expert qualified to interpret the results of the test performed on Davis’ blood samples. In summary, Dr. Valentour testified to the scientific basis of the blood test, its standard deviations, and in general the correlation between the blood alcohol concentration at the time of driving and the results of subsequent chemical tests. Specifically, Dr. Valentour testified that both the time and amount of alcohol consumption will affect the results of a subsequently administered test. In essence, he testified that the closer in time the test is performed to the time alcohol is consumed prior to or during driving the more accurately the test results will reflect the actual blood alcohol concentration at the time of driving. However, on this record we conclude that the trial court did not consider Dr. Valentour’s testimony beyond that which confirmed the accuracy of the blood test results at the time the tests were performed. Under the trial court’s interpretation of Code § 18.2-266(i), as we understand it, any evidence which tended to establish that the blood test results did not accurately reflect Davis’ blood alcohol concentration at the time he was driving was not considered relevant. For that reason, we will not on appeal, as suggested by the Commonwealth on brief, decide whether Dr. Valentour’s testimony adequately rebutted the rebuttable presumption established by the .10 percent blood test results. Rather, we hold that the trial court erred in its interpretation of Code § 18.2-266(i) and in not considering the relevant evidence to determine if Davis had offered evidence sufficient to rebut the presumption that the blood test results accurately reflected his blood alcohol concentration at the time he was driving his truck.

Finally, we note that Davis did not allege that he consumed alcohol after the accident. In fact, the evidence established that he did not. Accordingly, Davis’ reliance upon Overbee v. Commonwealth, 227 Va. 238, 315 S.E.2d 242 (1984) and Coffey v. Commonwealth, 202 Va. 185, 116 S.E.2d 257 (1960), where the evidence showed that the accused consumed alcohol after driving, and Fowlkes v. Commonwealth, 194 Va. 676, 74 S.E.2d 683 (1953) and Bland v. City of Richmond, 190 Va. 42, 55 S.E.2d 289 (1949), where the evidence failed to show that the accused *302did not drink after driving, is misplaced.

For these reasons the conviction below is reversed and the case is remanded to the trial court for retrial consistent with this opinion if the Commonwealth be so advised.

Reversed and remanded.

Coleman, J., concurred.

dissenting, but concurring in the remand for a new trial.

The statute under which Kenneth Davis was prosecuted required the Commonwealth to prove that Davis drove or operated his motor vehicle “while [having] a blood alcohol concentration of 0.10 percent or more by weight by volume as indicated by a chemical test administered in accordance with the provisions of § 18.2-268.” Code § 18.2-266(i). The majority and I agree that Code § 18.2-266 unambiguously proscribes driving while having a blood alcohol concentration of .1 percent or more by weight by volume.1 I also agree with the majority that the Commonwealth is required to prove those facts beyond a reasonable doubt. See Clemmer v. Commonwealth, 208 Va. 661, 666, 159 S.E.2d 664, 667 (1968). I disagree with the majority, however, concerning whether a presumption exists to aid the Commonwealth in proving a violation of Code § 18.2-266.

*303The majority reads into the statute a presumption which has the effect of relieving the Commonwealth of its burden of proof. I can find no basis upon which to conclude that the General Assembly intended or that the case law allows the creation of a presumption to absolve the Commonwealth of its burden of proving the elements of this offense. While I do not dispute that the General Assembly could have enacted a statute which provided a rebuttable presumption, it did not do so in Code § 18.2-266(i). When the General Assembly has intended to create a presumption, it has had no difficulty in making its intent clear. See, e.g., Code § 18.2-269 (“the amount of alcohol ... as indicated by a chemical analysis . . . shall give rise to . . . rebuttable presumptions”). I find nothing in either the text or logic of 18.2-266(i) that leads me to conclude the General Assembly intended anything other than that guilt would be proved beyond a reasonable doubt according to the normal processes of human reasoning and experience. Without any guidance as to legislative intent, I believe that the majority has simply re-written the statute. In my judgment, it simply defies human logic and experience to adopt a presumption, absent explicit legislature direction, that the results of an after-administered blood alcohol test provide an accurate measurement of blood alcohol level at an earlier point in time.

In discharging its burden of proving a violation of the statute, the Commonwealth must do more than simply offer proof of an after-administered chemical test showing a blood alcohol concentration of .1 percent or greater. The Commonwealth must prove that the accused had a blood alcohol concentration of .1 percent at the time he was operating his motor vehicle. Clemmer, 208 Va. at 664, 159 S.E.2d at 666 (“gravamen of the [intoxicant] offense [s] is driving while” in violation of the prohibition). Moreover, “the Commonwealth must establish both essential facts beyond a reasonable doubt.” Id. To prove its case under the statute, the Commonwealth should be required, through expert testimony or otherwise, to relate back the .1 percent reading from 10:17 p.m. (the time the test was administered) to “sometime prior to 9:00 p.m.” (the time of the accident). See State v. Rollins, 141 Vt. 105, 109, 444 A.2d 884, 886 (1982).

When the statute is viewed in this light, the evidence that Davis attempted to offer as proof concerning his condition at the time he was driving was relevant. Because the test result must be related *304back to the time of the driving, the trier of fact must be permitted to scrutinize evidence which is offered concerning the driver’s condition at the time of driving and any other relevant evidence probative of the driver’s blood alcohol level at the time of operation. Accordingly, Davis should have been permitted to offer evidence concerning his observable condition at the time of driving or upon his arrest.

[T]he defendant’s physical condition at the time of arrest was logically relevant. . . . “It can well be argued that if a person exhibited no physical signs of intoxication whatever but that the chemical test for alcohol shows a level above .10% that the test was inaccurate.” 1 R. Erwin, Defense of Drunk Driving Cases, § 16.05, at 16-25 (3d ed. 1981). Accord, State v. Clark, supra, 286 Or. at 39-40, 593 P.2d at 126-27; Denison v. Anchorage, 630 P.2d 1001, 1003 (Alaska Ct. App. 1981). The converse is equally true — observable symptoms support the accuracy of test findings. Most importantly, the evidence [might tend] to show that the alcohol level was elevated at the time of operation, not just at the time of the test. See State v. Swaregin, 12 Or. App. 290, 291, 506 P.2d 729, 729 (1973). This is a critical element of the prosecution’s case. Thus, evidence on the defendant’s drunkenness was probative on matters at issue in the case, and was properly admitted.

Rollins, 141 Vt. at 110, 444 A.2d at 887.

Because any evidence relevant to establishing whether the test reading of .1 percent at 10:17 p.m. accurately reflected Davis’s blood alcohol level at the time of the accident should have been admitted, the trial judge erred in excluding Davis’ evidence concerning his condition at the time of the accident. See People v. Kappas, 120 Ill. App. 3d. 123, 128, 458 N.E.2d 140, 143 (1983) (“Based on the totality of the circumstances, a jury could reasonably conclude that defendant’s [blood alcohol content] at the time he was driving was .10%”); Rollins, 141 Vt. at 110, 444 A.2d at 886-87 (“any evidence relevant to whether the [subsequent] reading accurately reflected the defendant’s blood alcohol level at [the time he was driving] was not only admissible, but indispensable to the . . . case”).

*305Of course, evidence concerning the symptoms of intoxication must be treated with care in a prosecution for driving with .10% or more alcohol in the blood. Prejudice to the defendant would result if the jury was unclear about the offense charged, and confused the .10% offense with driving under the influence. The trial court’s scrupulous instruction on the elements of the offense [must safeguard] . . . against any such prejudice in the trial. . . .

Rollins, 141 Vt. at 110, 444 A.2d at 887.

Thus, although I would reverse the conviction and remand for a new trial, I would not read into the statute a presumption and, thereby, relieve the Commonwealth of its burden to bring forth evidence to prove the requisite facts beyond a reasonable doubt.