GOODWIN, J.

Beverly Bland, plaintiff, appeals from a judgment of involuntary nonsuit granted in favor of the defendants Williams at the conclusion of her evidence in an action arising out of injuries sustained in an automobile accident.

Plaintiff was the guest passenger of Larry Williams, who was driving his father’s’ automobile at the time of the accident. While both Larry and his father were named as defendants, we will refer only to Larry herein. There was evidence that the car was new and that Larry was driving it for the first time.

*195 The sole question is whether there was sufficient evidence of gross negligence to permit the case to go to the jury. The trial court held that there was not.

Beverly and Larry were high school classmates and friends. With others of their approximate age, both had attended a dance in Carlton. After the dance, Beverly and three other friends accepted Larry’s offer of a ride to a restaurant in McMinnville. The time was about one a. m. The night was foggy. The road between Carlton and McMinnville is a state highway having a paved surface 18 feet wide with narrow shoulders on each side. The road had a number of curves and also followed the rolling topography by going up over rises and down into depressions.

The evidence that fog was present along the route of travel was not specific, but, giving effect, as we must, to the best evidence in the plaintiff’s case, there was evidence of sufficient fog in the atmosphere to limit visibility to a significant degree. Testimony concerning visibility varied from “50 feet in patches” to “about 200 feet.” Assuming that there were patches of fog in which visibility was restricted to 50 feet, the evidence indicated that such patches were frequent; that the visibility at the time and place of the accident may have been affected by fog; and that fog was a continuing factor which a prudent driver would have had in mind at the time.

The mechanics of the accident were described only sketchily by witnesses. As the car was proceeding along the road at a speed which the jury could have found to be 60 miles per hour, the passengers in the front seat were attempting to tune the car radio. None expressed any concern about speed. The driver, glancing momentarily away from the road, pushed a button which tuned the radio. At that instant, the plaintiff *196 called Ms name suddenly. Plaintiff recalled that she was prompted to call out the driver’s name because the car appeared to be going straight ahead while the road was curving gradually to the left. Almost immediately the witnesses heard the noise of gravel under the car.

During the fleeting moments when Larry noticed his difficulty and attempted to retain control, the vehicle apparently did not decrease speed. He said that the first thing he knew the car had edged off the pavement and at least one of the wheels on the right side had dropped into the ditch. Larry said he attempted to regain the pavement by turning the front wheels and accelerating the engine. This maneuver failed. Apparently, while the right wheels were in the ditch, the forward motion of the ear was abruptly altered when the ear struck the projecting end of a concrete culvert.

The testimony was understandably vague about what happened next, because the passengers were the only witnesses and the episode was concluded in an instant. Circumstantial evidence indicated that the car flipped upward, cutting a section out of a utility pole in a manner that left a stump seven or eight feet high. The car retained sufficient momentum to travel some 30 feet straight ahead into a driveway. This evidence was consistent with, and tended to corroborate, the other testimony concerning a speed of “between 50 and 60 miles per hour.” It did not of itself prove any specific speed. Navarra v. Jones, 178 Or 683, 169 P2d 584.

The negligence in the specific act which preceded the immediate loss of control was momentary inadvertence, and of itself would not constitute gross negligence under our decisions upon the subject. Kudrna *197 v. Adamski, 188 Or 896, 216 P2d 262, 16 ALR2d 1297; Williamson v. McKenna, 223 Or 366, 354 P2d 56.

The facts of the instant case are similar in many respects to those of Burghardt v. Olson, 223 Or 155, (349 P2d 792 adhered to) 354 P2d 871. In that case we held that traveling 60 miles per hour while approaching a curve was not evidence of recklessness under the test laid down in Williamson v. McKenna, supra. The critical distinction, of course, is the presence of fog in the case at bar.

The plaintiff urges that persistent speed in the face of the obvious hazards created by fog is an objective indication of a reckless state of mind despite contrary subjective evidence that Larry intended to be careful of his father’s new car.

This court has taken note of the effect of fog in cases involving ordinary negligence:

“Many collision cases have arisen where the vision was obscured by fog. Most courts hold that such weather conditions demand the exercise of a high degree of care but that it is not negligence as a matter of law to drive through fog: * * * [citing cases]. Any experienced driver knows that the utmost care must be exercised under such circumstances but that travel can be maintained with comparative safety when such care is exercised by those using the highway * * French v. Christner, 173 Or 158, 171, 135 P2d 464, 143 P2d 674.

The evidence of fog is essential in order to sustain plaintiff’s theory that the question of gross negligence should have gone to the jury. As we have pointed out, the evidence of fog was indefinite and speculative. It is established only that visibility might have been affected to some degree. It is equally established that *198 the speeds testified to could have been found to be ordinary negligence under the circumstances. It was not established that fog at the place where the accident occurred materially affected the control of the vehicle.

The presence of fog, therefore, does not simplify the problem in the instant case. Granting that there was negligence in operating the vehicle at the speed testified to, the trial judge was required to determine whether there was sufficient evidence of a reckless state of mind to bring the case within the strict requirements of the test laid down in Williamson v. Mc Kenna, supra.

Williamson v. McKenna adopted the rule found in Restatement, 2 Torts 1293, § 500:

“* * * Reckless Disregard of Safety Defined. “The actor’s conduct is in reckless disregard of the safety of another if he intentionally does an act or fails to do an act which it is his duty to the other to do, knowing or having reason to know of facts which would lead a reasonable man to realize that the actor’s conduct not only creates an unreasonable risk of bodily harm to the other but also involves a high degree of probability that substantial harm will result to him.”

Applied to the facts of this case, the rule quoted above bars the plaintiff unless we can say that intentionally driving at the speed testified to under the conditions existing amounted to evidence of a “reckless disregard” of the safety of others. We do not think the testimony justifies such a conclusion.

A reckless state of mind is rarely admitted by a defendant. In this case it is expressly contradicted by other evidence. In most of the guest-passenger cases, the quest for the state of mind of the host driver *199 involves working back from evidence of objective facts. For example, in Turner, Adm’r v. McCready et al, 190 Or 28, 222 P2d 1010, driving 50 miles per hour on a dusty gravel road with visibility reduced to 40 feet was held to be evidence of a reckless disregard for the safety of others when coupled with a failure to avoid a headon collision in or near the center of the traveled portion of the road.

In the instant case, the car actually ran off the road in a manner that could be explained under the evidence equally well with or without reference to the reduced visibility caused by fog. The driver momentarily lost control of his vehicle when he attempted to adjust the radio. There was- no evidence that the driver could not see the edge of the road. The evidence concerning lookout and control all centered upon the immediate event.

The evidence was deficient in connecting speed with the actual event that precipitated the difficulty. There is no evidence that the car could not have negotiated the curve safely if the driver had not glanced away from the road to adjust the radio. Nor does the evidence show that either speed or fog had anything to do with the right wheels going in the ditch. To work back from evidence of speed in order to find a reckless disregard for the rights of others in the instant case would permit the case to go to the jury on the basis of an abstract element of negligence which was not shown to be causally connected with the accident. There was no evidence of racing or “showing off”. There was merely evidence of the kind of heedlessness involved in most negligence cases.

The rule in Williamson v. McKenna precludes holding that there was evidence of recklessness when there was merely evidence of negligence. While the distinc *200 tion is difficult to draw, we held in Williamson v. McKenna that it must be drawn, and the formula for drawing the distinction was spelled out. A plaintiff must produce evidence that the defendant intentionally did an act (or failed to act) which resulted in a high degree of probability that harm would result.

In reviewing the evidence, the trial court correctly found that it fell short of the proof necessary.

Affirmed.