The issue in this prosecution of defendant for driving under the influence of intoxicating liquor, is whether the blood alcohol test result of . 12 percent is inadmissible because the blood sample on which the test was based was taken almost two hours after the collision.
The following is a time log of relevant events:
5:00-5:22 p.m. — Defendant drank four 12-oz. containers of beer, according to defendant’s testimony.
5:25-5.27 p.m. — Collision. While defendant was driving with a passenger on a township road, he crossed the highway and collided head-on with a utility pole. Defendant explained that his right front wheel just “dropped off’ the paved part of the road, causing him to cross over and strike the pole. Defendant’s passenger left the scene unidentified; defendant stated that he did not know the passenger’s name. No other vehicle was involved in the collision.
6:30 p.m. — The investigating officer arrived and observed defendant as he was seated in the ambulance; defendant’s speech was slurred, mushy and incoherent, and he had an odor of beer upon him.
7:15 p.m. — Blood sample drawn at hospital.
Defendant’s objection to admissibility of the test the result was taken under advisement in order to study material submitted by defense counsel in support of the objection. We now overrule the objection, and admit the offered test result.
Defendant is charged with operating a motor vehicle in violation of subsections (a)(1) and (a)(4) of Vehicle Code §3731, 75 Pa.C.S., which provide as follows:
“(a) Offense Defined. — A person shall not drive, *39operate or be in actual physical control of the movement of any vehicle while:
(1) under the influence of alcohol to a degree which renders the person incapable of safe driving;
(2) . . .;
(3) . . . ; or
(4) the amount of alcohol by weight in the blood of the person is 0.10% or greater.”
The evidence supports guilt under subsection (a)(1) but not under (a)(4).
The material submitted for our study consists of a law review article entitled: “The Single Chemical Test For Intoxication: A Challenge To Admissibility”, by Edward F. Fitzgerald and Dr. David N. Hume, 66 Massachusetts Law Review, No. 1, Winter, 1981. It points out the speculative character of blood alcohol test results taken at a substantial interval of time after the driving ceased, depending on the amount of time required for the ingestion of consumed alcohol into the blood stream (absorption period) and the time required for dissipation (clearance period) of the alcohol from the blood stream after reaching its peak of influence following absorption, posing the chance that a person, after consumption of alcohol, may be found to be under the influence at the time of testing during the latter part of the absorption period or the early part of the clearance period, yet not under the influence at the time of driving during the early part of the absorption period. The authors state (Part IV, C and D):
“C. Usual ‘Reasonable Time’ Assumption Not Valid
“The ‘reasonable time’ test as presently applied needs careful reevaluation. The concept that a test within two hours after an incident is, somehow, inherently reliable must be abandoned. The problem of interpretation of a later test exists no matter what *40type of intoxication test is involved. It arises whenever there is significant delay between the time of offense and the time of test or sample collection. A test thirty minutes or longer after operation is not within a ‘reasonable time’ unless the result reflects the ‘time of offense’ BAC with reasonable certainty.
“A test for BAC at a time so close to the time of offense that any rise or fall in the BAC in the interim would be inconsequential would be a test within ‘a reasonable time.’ A sample within ten or fifteen minutes after the incident would probably meet that qualification.
“A sample within fifteen to thirty minutes becomes more questionable unless a declining BAC during that interval could be confirmed by some method, thus supporting a conclusion that the earlier BAC was as high or slightly higher. The present state of the art, however, does not usually lend itself to confirmation of a rising or falling BAC at the time of the test.
“D. A Second Test — Prerequisite for Admissibility
“A second chemical test for intoxication or even a series of such tests will not solve the essential problem, as long as there still remains a significant delay between the time of offense and the first sample collected.
“However, a second test, perhaps one-half hour after the first, ought to be required by statute when chemical tests are to be used in crimfiafil cases. In some cases the second test will support the prosecution. In many, the significance of the two tests might still remain ambiguous. In others, the second will be exculpatory, of particular importance where the single test alone would have been inculpatory. Where ‘implied consent’ laws provide for loss of license upon refusal to submit to such tests, the *41even-handed administration of justice should mandate that such a second test be required.”
The authors further recommend development of devices for testing “at-the-scene”, the development of a procedure for determining whether the blood alcohol concentration (BAC) is rising or declining at the time of the test (Part IV, E and F), and a procedure for a second test “perhaps one-half hour after the first” which could be decisive in some cases (Part IV, D).
The relevance of the duration of the absorption period, the time when the peak of influence is reached, and the duration of the clearance period, in relation to the times of driving and consumption of alcoholic beverage, was judicially recognized in Commonwealth v. Hartman, 179 Pa. Super. 134, 119 A.2d 21 (1956) reversed 383 Pa. 461 (1956), in Commonwealth v. Supanick, 26 Somerset L. J. 367 (1971) and cases cited therein, and in Commonwealth v. King, 32 Somerset L. J. 81 (1976). In King, we held that the result of a test made more than two hours after driving was inadmissible, where there was no direct evidence of symptomatic behavior observed at or near the time of driving, nor any evidence establishing the times of the absorption and clearance periods; a new trial was granted for want of sufficient evidence of intoxication at the time of driving. Accord: Commonwealth v. Martz, 34 Somerset L. J. 25 (1977); see also Commonwealth v. Funk, 254 Pa. Super. 233, 385 A.2d 995 (1978). In Supanick, the conviction was upheld on the basis of direct police observation of defendant’s symptomatic behavior at the time of driving and their opinion testimony that defendant was then under the influence, and the breathalyzer test taken one-half hour later. In Commonwealth v. Rounds, 38 Somerset L. J. 16 (1979), the breath test was giv*42en almost two hours after drinking, and there was additional evidence of observed symptoms of defendant’s being under the influence at the time of driving, and the conviction was upheld. In all of our cases, there was no consumption of alcoholic beverage after driving. Thus our holdings seem consistent with the two-hour test limit rule referred to by Fitzgerald and Hume quoted above. See also Gary v. Skipalis, 69 Luzerne 106 (1979). Under the foregoing decisions, the instant test taken just short of two hours after the driving is admissible; moreover, this defendant’s behavior symptomatic of being under the influence was observed one hour after driving.
Hence, the test result is admissible here within the above authorities.
Later Superior Court decisions have taken a more liberal attitude than King and Martz, both supra, toward admissibility of test results made hours after driving. That court has adopted the rule that since the statute places no time limit within which the blood test must be taken, the court should not do so, and that the interval of delay between those events is a matter of evidentiary weight for the fact finder, not a matter of admissibility. Thus, in Couts v. Ghion, 281 Pa. Super. 135, 421 A.2d 1184 (1980), a civil action for wrongful death brought against a liquor licensee, in which the issue was whether the vehicle driver was visibly intoxicated when served his last drink at 5:15 p.m., the evidence showed that thereafter the driver drove his car until 6:00 p.m. when the accident happened; his car had suddenly crossed the double yellow divider line, proceeded into the opposite lane for 300-400 feet where he collided with decedent’s vehicle, causing the death. Defendant driver did not apply his brakes, crossed back over into his own lanes, ran off the roadway and ultimately ran off the road into a cemetery. Both *43vehicles were extensively damaged, defendant driver was operating in the darkness without lights, and within minutes after the accident he was observed by a police officer with clear symptoms of then being under the influence. A breathalyzer test was administered at 8:31 p.m., about two and a half hours after the accident and more than three hours after he had his last drink at defendant licensee’s bar. The test result was .12 percent, and at 8:43 p.m. a blood test gave the same result. Of course, under Supanick, King and Martz, there was ample evidence of intoxication at the time of driving of which the later test results were corroborative; but the issue was whether the test results at 8:31 and 8:43 p.m. were admissible on the question of whether defendant driver was visibly intoxicated when he had his last drink at 5:15 p.m., over three hours earlier, of which there was no direct evidence. The court affirmed admission of the test results over defense objection. The court said that “. . . evidence of elevated blood alcohol content, zuhen accompanied by other evidence suggesting a high degree of intoxication, is admissible” (144 Emphasis added), that the facts in the case, “together with evidence of the results of the breathalyzer and blood tests” (145), were sufficient to support the verdict for plaintiff, and that the “test results are relevant circumstantial evidence bearing on the question of whether Ghion was visibly intoxicated when he was served. . .” (145). Judge Van DerVoort dissented.
Analogizing the officer’s observations in Ghion 45 minutes after the last drink, to the officer’s observations here about one hour after the collision, and further analogizing the blood alcohol test in Ghion three and one-fourth hours after the last drink and two and one-half hours after the officer’s initial observation, to the test here almost two hours after ob*44servation, the conclusion of admissibility in Ghion tends to support admissibility here. As pointed out in footnote 6 of Ghion, the differences in the cases bear on the weight of the evidence, not admissibility. Compare Cusatis v. Reichert, 267 Pa. Super. 247, 406 A.2d 787 (1979), discussed in Ghion.
The fact that Ghion is a civil rather than a criminal case is not decisive of the issue of admissibility. In Commonwealth v. Tylwalk, 258 Pa. Super. 506, 393 A.2d 473 (1978), a prosecution for involuntary manslaughter and driving under the influence, a hit-run collision occurred at 12:15 a.m.; defendant had arrived home 15 or 20 minutes later and was found there, in bed, about 1:15 a.m. He was taken for blood alcohol testing, but no test was performed until 4:30 a.m. The report of the case in the opinion mentions no symptomatic or other evidence of intoxication, other than the test result which the trial judge had suppressed as being too remote in time and place from the accident. The Commonwealth appealed and the Superior Court reversed the suppression order, holding that since the statute established no time limit for testing, the test was admissible and that the effect of the passage of time was for the trier of fact to consider. Compare: Commonwealth v. Trefry, 249 Pa. Super. 117, 375 A.2d 786 (1977), test one and one-half hours after collision, and symptomatic observations; Commonwealth v. Arizini, 277 Pa. Super. 27, 419 A.2d, 642 (1980) more than two hours.
None of the above cases involves the 1982 amendment to §3731 of the 1976 Vehicle Code, subsections (a)(1) and (4) of which are here involved and are quoted ante. We should briefly consider their interrelationship. Subsection (a)(1) requires that defendant be “under the influence of alcohol to a degree which renders the person in*45capable of safe driving”, and subsection (a)(4) requires only a blood alcohol concentration of .10 percent or greater. Obviously, a test result of .10 percent or greater of itself renders the person incapable of safe driving; as stated in Schwartzbach v. Dunn, 252 Pa. Super. 454, 462, 381 A.2d 1295 (1977):
“A blood alcohol content of .10 or above raises a presumption of intoxication. We feel that this presumption implies that a person with a blood alcohol content of .10 or above is unfit to drive. Otherwise why would driving under such conditions be unlawful?”
Compare Commonwealth v. Giehll, 42 Somerset L. J. 271 (1983). In Commonwealth v. Thompson, 27 Somerset L. J. 241, 246 et seq. (1971), affirmed per curiam 221 Pa. Super. 824, 292 A.2d 425 (1972) allocatur denied, we said (248), quoting from State v. Johnson, 199 A.2d 809, 822 (N.J. 1965):
“ . . no one has ever been found with that quantity [. 10%] of alcohol in his blood who was not so seriously impaired in his faculties, irrespective of external manifestation, that he should not operate a motor vehicle.’ ”
In Commonwealth v. DiFrancesco, 458 Pa. 188, 198, 329 A.2d 204 (1974) the court said:
“In its deliberations prior to enacting §624.1 [predecessor of §3731] the legislature considered evidence demonstrating that, in virtually all drivers, driving ability is significantly affected by a blood alcohol content of 0.10% or more. No evidence to the contrary was introduced at the trials below, and our research suggests that no persuasive evidence of this sort is presently available.”1
*46From the foregoing it follows that if defendant had a test result of. 10 percent or more under (a)(4) at the time of driving, he was also logically in violation of (a)(1), but the reverse is not true. Nevertheless, although the cases admit in evidence for consideration by the fact finder test results distant in time from the time of driving on the issue of capability of safe driving, none as yet states that such remoteness is irrelevant under (a)(4); we think that a constitutional construction of (a)(4) mandates proof that the .10 percent blood alcohol concentration must be shown to have existed at the time of driving. Compare King, supra, 84-85. This does not mean that the test must be given at the time of driving, nor require apprehension and observation at that moment, nor use of a portable testing device (see Commonwealth v. Quarles, 229 Pa. Super. 363, 386, 324 A.2d 452 (1974). It is possible to calculate, from a given blood alcohol level ascertained as of the time subsequent to driving, the probable alcohol level at the time of prior operation, at the average rate of .015 percent per hour at which alcohol clears after the clearance period begins, provided that the driving occurred during the clearance period; and it is possible to establish when the condition of under-the-influence was reached and when the clearance period began, given the times of drinking and testing. See King, supra, 86 note [4] and 88 note [6], Expert testimony is essential, however, to establish those relationships. Commonwealth v. Guiliano, 274 Pa. Super. 419, 418 A.2d 476 (1980); compare McCormick On Evidence (2nd Edition, West, 1972) *47§209. No such expert testimony was produced here.2 Thus, it seems to us that, unless the blood alcohol test is taken at or very near the time of driving, there must be evidence, in addition to a later test result, establishing that the blood alcohol level was at least .10 percent at the time of driving in order to find guilt under subsection (a)(4). Here, we cannot say that defendant, who had a test reading of .12 percent one hour and 45 minutes after driving, also had a .10 percent when driving.3 Accordingly, the evidence is insufficient to find defendant guilty under (a)(4), but we can and do find him guilty under (a)(1). We also note that we need not say in this case just how near to the time of driving the test must be in order to convict under (a)(4) without expert testimony.4 •
*48NONJURY VERDICT
Now, May 3, 1984 we find defendant guilty as charged under subsection (a)(1), not guilty under subsection (a)(4), of Vehicle Code §3731. Unless a timely post-trial motion is filed, defendant shall be scheduled for sentencing sec reg.
. Note 3 of the opinion in DiFrancesco supra cites Thompson supra with approval.
*46See also: Emerick v. Carson, Pa. Super. , 472 A.2d 1133 (1984), note 1, and Commonwealth v. Seville, 266 Pa. Super. 587, 405 A.2d 1262 (1979), stating that blood alcohol tests are accurate and that the results are medical facts.
. In Commonwealth v. Daugherty, 259 Pa. Super. 88, 393 A.2d 730 (1978), the court approved a statement in the trial judge’s instructions to the jury that “people pass alcohol out of their systems and that one test the jury may use in determining when consumption occurred is whether the person stays the same, improves or gets worse during the time he is under police observation,” without stating any particular rate of passage, on the ground that such passage is a matter of common knowledge and may be taken judicial notice of.
. Some authorities fix the average absorption period at 45 minutes to one hour and 15 minutes (see King note [4]); Fitzgerald and Hume supra say it may be as low as 15 to 30 minutes, depending on stomach contents and character of liquor consumed (Part II). In this case, defendant may or may not have reached the peak of influence at time of collision in relation to the time of drinking. If he had reached or passed that peak, relating the . 12 percent result backwards at .015 percent per hour (King, note [6]), based on an interval of one and three-fourth hours between collision and test, would produce a blood alcohol concentration of .09375 at the time of collision.
. See Commonwealth v. Mitchell, (No. 510 Criminal 1983, slip opinion of May 7, 1984, page 6).