OPINION PAUL PRESSLER, Justice. Appellant was convicted of driving while ■ intoxicated. The court assessed punishment at one year in jail, probated for two years, and a $500 fine. Appellant complains that his consent to take an intoxilyzer test was involuntarily given. We reverse and remand for a new trial. A police officer saw the appellant driving in circles in a parking lot. The officer stopped him and, after observing him, gave him the warnings from the implied consent law. See TEX.REV.CIV.STAT. art. 6701Z -5. Appellant agreed to take a breath test.

The intoxilyzer indicated a blood alcohol content of 0.16 percent. The first point of error is that the trial court should have suppressed the results of the test. Appellant alleges that his consent was involuntary because the officer incorrectly told him that the implied consent law applied. This statute applies only to driving on a public highway or a public beach but not to driving on a parking lot. See art. 6701l-5. On the other hand, the driving while intoxicated law does apply to parking lots. See art. 6701l-1. In essence, appellant has found a loophole which only legislative action can close.

A reversal is required. Hall v. State, 649 S.W.2d 627 (Tex.Cr.App.1983); Turpin v. State, 606 S.W.2d 907 (Tex.Cr.App.1980); see TEX.CODE CRIM.PROC. art. 38.23. The first point of error is sustained. The second and third points of error are not reached because they presuppose the overruling of the initial point of error. The fourth point of error relates to testimony of the arresting officer about a sobriety test’s being given to the appellant. The officer performed a horizontal gaze nystagmus (HGN) test before taking the appellant to the station.

The HGN test calls for the subject’s eyes to follow the movement of an object. As the object moves steadily to one side of the subject’s field of vision, the subject’s eyes eventually fail to track the object smoothly. The HGN test presumes that a sober person will exhibit smooth eye movement up to a greater angle than an intoxicated person. Appellant contends that since the officer was not qualified as an expert, the evidence of the HGN performance should not have been admitted. The state maintains that an HGN test is merely an optical version of the routine “walk a straight line” test.

No Texas case addresses the use of HGN evidence in DWI cases, but other states have considered the issue. The most persuasive decision is that of the Arizona Supreme Court in State v. Superior Court, 149 Ariz. 269, 718 P.2d 171 (1986). That case holds that HGN evidence is proper as to the issue of intoxication but not as to precise blood alcohol content. In other words, the HGN results are admissible for qualitative — but not quantitative— purposes. Contra People v. Loomis, 203 Cal.Rptr. 767, 156 Cal.App.3d Supp. 1 (1984). Any lay witness may give an opinion as to intoxication. See TEX.R.CRIM.EVID. 701.

The fourth point of error is overruled. See generally Frye v. United States, 293 F. 1013 (D.C.Cir.1923) (addressing standards for permitting evidence of a scientific test); People v. Vega, 145 Ill.App.3d 996, 99 Ill.Dec. 808, 496 N.E.2d 501 (1986) (reversing a DWI conviction because of an inadequate predicate for testimony about an HGN test). The conviction is reversed and the cause remanded for a new trial.