I respectfully dissent.
The trial court’s task in this case was to determine, using its independent judgment, whether the administrative decision was supported by the weight of the evidence. (Santos v. Department of Motor Vehicles (1992) 5 Cal.App.4th 537, 545 [7 Cal.Rptr.2d 10].) On appeal, that determination will be reversed only if it is not supported by substantial evidence (ibid.; Coombs v. Pierce (1991) 1 Cal.App.4th 568, 576 [2 Cal.Rptr.2d 249]) or is based upon an erroneous conclusion of law (Webb v. Miller (1986) 187 Cal.App.3d 619, 625 [232 Cal.Rptr. 50]).
In the instant case, the trial court ruled the evidence used by the Department of Motor Vehicles (DMV) to support its finding Yordamlis was driving a motor vehicle when he had a blood-alcohol content (BAC) of 0.08 percent or more was insufficient “as a matter of law.” The trial court based its ruling on (1) lack of foundation for the admissibility of the BAC test results, (2) lack of foundation as to the qualifications of the person who performed the chemical analysis of Yordamlis’s blood test, and (3) failure to demonstrate a relationship between the BAC test results and Yordamlis’s BAC at the time of driving. Since the court’s ruling, however, appellate court decisions have *664established that the first two bases are erroneous as a matter of law. (Davenport v. Department of Motor Vehicles (1992) 6 Cal.App.4th 133, 140-143 [7 Cal.Rptr.2d 818]; Santos v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 547; McKinney v. Department of Motor Vehicles (1992) 5 Cal.App.4th 519, 525-526 [7 Cal.Rptr.2d 18]; Burge v. Department of Motor Vehicles (1992) 5 Cal.App.4th 384, 389 [7 Cal.Rptr.2d 5]; Imachi v. Department of Motor Vehicles (1992) 2 Cal.App.4th 809, 816-817 [3 Cal.Rptr.2d 478].) Consequently, the only remaining basis for the court’s ruling is failure of the DMV to offer expert opinion connecting Yordamlis’s BAC of 0.17 percent at the time of the test to his BAC at the time of his driving.
Because the trial court ruled the above evidence insufficient as a matter of law, this court is not, as my colleagues assume, limited to substantial evidence review, but, rather, may reach its independent determination based on the undisputed facts. (E.g., Webb v. Miller, supra, 187 Cal.App.3d at p. 625.) Based on the undisputed facts, I believe the trial court erred in finding the evidence insufficient as a matter of law to support suspension of Yordamlis’s driver’s license.
The law is settled that BAC at time of driving can be shown only by circumstantial evidence and that a chemical test result is merely one kind of circumstantial evidence. (Burg v. Municipal Court (1983) 35 Cal.3d 257, 266, fn. 10 [198 Cal.Rptr. 145, 673 P.2d 732]; McKinney v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 526, fn. 6; People v. Randolph (1989) 213 Cal.App.3d Supp. 1, 6-7 [262 Cal.Rptr. 378].) In the instant case, the hearing officer had before him the following circumstantial evidence: (1) the sworn statement of Officer Bhatnagar, the arresting officer, that Yordamlis was stopped because he was driving erratically and, at the time of the stop, his eyes were bloodshot and watery, his speech slurred, his gait unsteady, and he had about his person an odor of alcohol; and (2) the report of the blood test results showing that at the time Yordamlis’s blood was drawn he had a BAC of 0.17 percent. In addition, the hearing officer was permitted to draw inferences and deductions from the evidence before him. (Evid. Code, § 600, subd. (b); compare McKinney v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 524 [absent contrary evidence, hearing officer could infer police officer observed driver driving in an inebriated condition shortly before his arrest] with Santos v. Department of Motor Vehicles, supra, 5 Cal.App.4th at pp. 549-550 & fn. 9 [given contrary evidence, hearing officer could not infer driver’s BAC at time of driving was at least as high as at time of testing].) Here, the evidence showed objective indicia of intoxication at the time of Yordamlis’s arrest and a BAC of more than double the legal limit.
Pursuant to Vehicle Code section 23152, subdivision (b), if the blood sample was collected within three hours of Yordamlis’s arrest, the hearing *665officer, absent contrary evidence, would be required to find the test results established Yordamlis’s BAC at time of driving. (Burge v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 391; accord, Bell v. Department of Motor Vehicles, ante, p. 304 [13 Cal.Rptr.2d 830].) If, by contrast, the sample was collected anytime after three hours, the hearing officer, in my view, could reasonably infer that at the time of driving Yordamlis’s BAC was at least equal to and probably higher than the 0.17 percent shown by the test.1
The logic of the latter inference has uniformly been recognized by case law in this state. Thus, in In re Martin (1962) 58 Cal.2d 509, 512 [24 Cal.Rptr. 833, 374 R2d 801], our Supreme Court stated: “It is a matter of common knowledge that the intoxicating effect of alcohol diminishes with the passage of time; hence, the probative value of a blood test diminishes as well. In a short period of time an intoxicated person may ‘sober up’ sufficiently to negate the materiality of a blood test where the sample has not been timely withdrawn. [Citation.]” (Accord, People v. Schrieber (1975) 45 Cal.App.3d 917, 922 [119 Cal.Rptr. 812] [“[T]he greater the elapse of time the greater the benefit to the defendant, as the alcohol in his system becomes dissipated by the normal body processes.”]; see also Santos v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 549, fn. 8 [“If the blood was drawn more than three hours after the arrest, the test results would reflect a blood-alcohol level lower than the maximum that had been reached before the alcohol began to dissipate”].) Consequently, regardless of when the blood sample was taken, the 0.17 percent test result, together with the other evidence, was sufficient to permit the hearing officer to find the DMV had met its burden of proof that Yordamlis was driving with 0.08 percent or more BAC.
Santos v. Department of Motor Vehicles, supra, does not compel a contrary conclusion. In Santos, as here, the DMV introduced the arresting officer’s sworn statement showing probable cause to arrest the driver and that she displayed objective indicia of intoxication, together with a blood test report showing, in that case, a BAC of 0.13 percent, with no indication of the time the sample was taken. The Santos court expressly recognized that (1) the untimed test results were admissible; (2) the hearing officer could rely on reasonable inferences and deductions from the evidence before him or her; and (3) if the blood was drawn more than three hours after the arrest, the test results would reflect a BAC lower than the maximum that had been reached before the alcohol began to dissipate. (5 Cal.App.4th at pp. 542, 549 & fin. 8.) In contradistinction to this case, however, the driver in Santos introduced *666evidence that her BAG would have been less than 0.08 percent at the time of driving “because at least the last drink she consumed would not yet have been absorbed.” (Id. at p. 548.) In light of that evidence, the Santos court correctly held that the time the blood sample was drawn was “a piece of information critical to a determination” of the driver’s BAG at the time of driving. (Id. at p. 549.) Lacking such information, “[t]he Department did not rebut respondent’s evidence that the test results failed to prove her blood-alcohol level at the time of driving was 0.08 percent or higher” (ibid., italics in original) and thus failed to meet its burden of proof.
Here, of course, Yordamlis introduced no evidence contrary to the DMV’s initial showing. The superior court thus erred in ruling the DMV failed as a matter of law in its burden of proof. (Cf. McKinney v. Department of Motor Vehicles, supra, 5 Cal.App.4th at p. 526, fn. 6 [absent contrary evidence, circumstantial evidence driver was intoxicated when driving provided independent support for hearing officer’s determination].)
In concluding the blood-alcohol test results were probative circumstantial evidence of Yordamlis’s BAG at the time of driving, I in no way intend to condone the deficient procedures in this case that permitted collection of a blood sample with no indication of the time the sample was collected. Surely, it would be a simple matter to devise a specimen label or form that requires notation of the time a sample is collected, as, indeed, the DMV evidently has already done. I merely advance the unremarkable proposition that omission of the time here did not necessarily deprive the test results of probative value as a matter of law.
I would reverse.
A petition for a rehearing was denied January 5, 1993.
The inference, of course, is permissive, not required. But if the hearing officer in the exercise of his discretion draws the inference, he does not, as the superior court ruled, err as a matter of law in relying thereon in determining the DMV has made a prima facie case.