On February 15, 1973, defendant was convicted by a jury of driving under the influence of intoxicating liquor. At the trial, the Commonwealth introduced into evidence the *274results of a breathalyzer test given to defendant. The testimony of the operator of the breathalyzer, Trooper Seiler, of the Pennsylvania State Police, indicated that part of the procedure followed after giving the actual test to the suspect was to run a simulator test with a .10 percent known solution of alcohol. On cross-examination, it was revealed that in this case the simulator test produced a reading on the machine of .08 percent instead of .10 percent. Trooper Seiler explained that the known solution disintegrated over a period of time and often resulted in lower readings. He also indicated that a precise result on the simulator test was not important.
At the close of the Commonwealth’s case, defendant moved to exclude the result of the breathalyzer because the incorrect result on the simulator test displayed that the test was inaccurate or, alternatively, that the machine was not shown to be accurate. The motion was overruled. Defendant has subsequently filed a motion in arrest of judgment and for a new trial contending that his motion to exclude the result was improperly overruled.
The basis of the admissibility of this evidence is set forth in The Vehicle Code of April 29, 1959, P. L. 58, sec. 624.1, as amended, 75 PS §624.1(b), which provides that the results of breath tests conducted on equipment approved by the Secretary of Revenue and administered by qualified personnel are admissible as evidence of intoxication. The breathalyzer has been approved by the secretary and case law affirms the admissibility of the results: Commonwealth v. Kaufold, 14 Lebanon 45 (1971), affirmed 222 Pa. Superior Ct. 275 (1972).
Pennsylvania courts have not, however, ruled upon the criterion for the admissibility of chemical tests. We have, therefore, examined the case law of other *275States. The vast majority of courts which have considered the problem have required the prosecution to offer evidence that (1) the operator was qualified to administer the test; (2) the test was administered properly; and (3) the machine was working properly, before the results may be considered by the jury. See State v. Lanahan, 110 N.J. Super. 578, 266 A. 2d 323 (1970); State v. Baker, 56 Wash. 2d 846, 355 P. 2d 806 (1960); State v. Woodward, 1 Oregon App. 338, 462 P. 2d 685 (1969); State v. Hood, 184 S.E. 2d 334 (W. Va., 1971); State v. Quinn, 289 Minn. 184, 182 N.W. 2d 843 (1971); State v. Sickles, 25 Ohio App. 2d 1, 265 N.E. 2d 787 (1970); People v. Donaldson, 319 N.Y. Supp. 2d 172, 36 A. D. 2d 37 (1971); Pruitt v. State, 216 Tenn. 686, 393 S.W. 2d 747 (1965); State v. Magoon, 264 A. 2d 779 (Vt., 1970). Since both men and machines are known to be fallible, it seems that it should be incumbent upon the prosecution to offer prima facie evidence establishing these criteria before the results of the breathalyzer are admissible in evidence.
In this case, the Commonwealth did offer evidence as to the qualification of the operator and administration of the test which are not challenged as incomplete or incorrect. Thus, we need not examine the evidence which established the first two criteria. The sole issue presented is whether the Commonwealth offered sufficient evidence of the proper working of the breathalyzer to have the results submitted to the jury.
On direct examination, Trooper Seiler testified to the preliminary steps which he followed in determining that the machine was operating correctly. He then explained briefly the simulator test with a known solution of alcohol which was apparently part of standard procedure followed after the breathalyzer test is given to the suspect. It was this simulator test *276which resulted in a reading on the machine .02 percent lower than it should have.
Trooper Seiler, however, as an operator of the machine, did not purport to understand the technical and chemical intricacies of the breathalyzer. Thus, while he was qualified to give prima facie evidence that the machine was working properly (State v. Magoon, supra; State v. Quinn, supra) he was not qualified to explain the discrepancy in the result of the simulator test. Trooper Seiler obviously did not qualify to give expert testimony on the decomposition of the test solution.
Some information on the simulator test is available, and it is apparent from an analysis of this information that the test is performed to check the accuracy of the machine. Perhaps the best statement on this standardization procedure is contained in Determining the Competence of a Breathalyzer Operator, by J. D. Taylor, Ph.D., Department of Pharmacology, University of Alberta, 9 Canadian Bar Journal 299 (1966), at page 304, where it is stated:
“ [T] he operator should be asked whether he checked the operation of the breathalyzer by using a ‘simulator or ‘artificial drunk’ containing a standard alcohol solution. The standard alcohol solution should not be confused with the potassium dichromate reagent which is contained in the ampoules. The ampoules are used in the determination of the blank, the standardization and the actual assay whereas the standard alcohol solution is used only during the standardization. The ‘simulator’ or ‘artificial drunk’ is a container wherein air can be bubbled through the standard alcohol solution and the alcohol laden air is then used in place of the breath sample. This is done to check all the working components such as the volume of air in the cylinder, the composition of the *277test reagent, the calibration of the instrument, the photoelectric cells and electrical system. If this standardization is correct then the instrument is in proper operating condition and is giving the correct answers for the samples presented for assay.” (Italics supplied.)
Also see State v. Sickles, supra, 265 N.E. 2d at 790; Howes, Hallatt, and Lucas, A Study of the Accuracy of the Breathalyzer as Operated by Police Personnel, 12 Journal of Forensic Sciences 444 (1967); Gottlieb, The South Carolina Implied Consent Law: The “Breathalyzer” and the Bar, 22 South Carolina Law Review 195 (1970).
Since the simulator test was used to' determine the accuracy of the machine, it should have produced a correct reading. When it did not, it became necessary for the Commonwealth to explain the discrepancy by expert testimony. Since the prosecutor did not offer expert testimony to prove that the breathalyzer was working properly, the results of the test were improperly admitted as evidence of intoxication.
Since defendant was admittedly driving, the sole issue was whether he was at the time under the influence of alcohol. Although there was other testimony which could have supported the verdict, the result of the breathalyzer test administered to defendant was .10 percent and created a statutory presumption of intoxication: The Vehicle Code, supra, section 624.1(c)(3). It was, therefore, prejudicial error to allow the jury to consider this evidence, and a new trial must be granted.
ORDER OF COURT
And now, July 10, 1973, for the reasons set forth above, the motion in arrest of judgment is denied. The motion for new trial is granted and the district attorney is directed to relist the case for trial at the next term of court.