PER CURIAM.

In this products liability action, plaintiffs appeal from the grant of summary disposition to defendants, following the trial court's decision to strike the testimony of plaintiffs' only expert witness. We affirm.

Plaintiff Danny R. Green (hereafter plaintiff) was the owner of a 1986 Ford pick-up truck that was involved in an accident in the early hours of New Year's Day, 1987. A friend of plaintiff's, Greg Hornfischer, was driving the truck; plaintiff was asleep in the passenger seat.

The accident occurred on a two-lane, paved highway in Ohio as Hornfischer and plaintiff were en route to visit plaintiff's sister. Hornfischer testified that he was driving approximately forty miles per hour when he saw a sign warning of a curve ahead and reducing the speed limit to thirty-five miles per hour. The road, in fact, curved sharply to the left, at an angle of approximately ninety degrees.

Hornfischer testified that he applied the brakes when he saw the sign, at a distance variously estimated at fifty, one hundred, or two hundred feet before the curve. After he stepped on the brakes "everything went crazy." The truck "started shaking and shuttering." Hornfischer was unable to recall what happened next: the truck went off the road, flipped over at least once, and landed on its wheels in a ditch that ran parallel to the road. Both plaintiff and Hornfischer were injured.

The accident was investigated by a deputy sheriff, Officer Miller. He inspected the road surface without finding any tire marks "on the roadway prior to its [the truck's] entering the ditch." He then issued a citation to Hornfischer for failure to control the vehicle, because "his stated speed contributed to the cause of the accident." Two of plaintiff's relatives, John Bukovich and Gail Bukovich, also inspected the accident scene. They both observed some skid marks on the road and the shoulder, either before the point at which the road curved or "a little ways into" the curve. No one observed any skid marks "in" the curve itself.

The truck pulled from the ditch soon after the accident by a tow truck operator, Dale Bloomfield. Bloomfield found that the truck's wheels were locked when he attempted to tow it out of the ditch. He was able to free the rear wheels by running the truck back and forth for a short distance on the highway. He testified that the rear wheels were freed simultaneously, although he was not able to observe the right rear wheel himself from inside the cab of the wrecker. Plaintiff's nephew, Gail Bukovich, who arrived at the scene just before the towing operation began, testified that he saw one rear wheel break free first and the other break free during the second attempt.

The truck was returned to Michigan, where plaintiff filed suit in the Macomb Circuit Court, naming Ford Motor Company, the manufacturer of the truck, and Jerome-Duncan, Inc., the dealership that had sold him the truck, as defendants. Plaintiff's theory was that a defect in the design of the truck had caused the brakes to lock when Hornfischer applied them as he approached the curve. The lock-up allegedly caused the truck to skid, run off the road, and flip over.

Plaintiff engaged the services of an expert witness, Dr. Rudolph Limpert, an automotive engineer with special expertise in brakes and braking systems. Plaintiff deposed Dr. Limpert on January 24, 1989; the deposition was videotaped for use at trial. Dr. Limpert had not inspected the accident scene in person and based his opinion on deposition testimony (Hornfischer, both Bukovichs, plaintiff, Bloomfield, and Officer Miller), photographs, the police report, and the truck itself.

After some calculation, Dr. Limpert determined that the lock-up and skid happened after the truck had entered the curve. He estimated that the earliest point where the lock-up could have occurred was "thirty to sixty feet" before the end of the curved portion of the road, "approximately three-quarters of the way through the curve."

Dr. Limpert's conclusion was that the brakes had "locked up" when Hornfischer pressed the brake pedal, causing the skid and the accident. He testified that when the truck was only "lightly" loaded (the driver, one passenger and no cargo would be a "light" load), applying the brakes would send too much hydraulic pressure to the rear wheels, causing them to lock. This problem could have been avoided, he said, if the manufacturer had incorporated a "proportioning" valve into the brake system. The valve would have automatically adjusted the pressure transmitted to the rear wheels according to the weight on the rear axle, preventing the brakes from locking and causing a skid.

During the deposition, defense counsel questioned Dr. Limpert about some of his conclusions, which were inconsistent with portions of the factual evidence that had previously been obtained. Dr. Limpert's responses essentially discounted the fact witnesses' testimony. On one point, he said that Hornfischer "does not know what happened" when he applied the brakes. On another, he agreed that his theory of the accident would predict that both rear wheels, having locked up together, would be freed at the same time; Gail Bukovich, however, had seen them break free individually. Dr. Limpert asserted that he did not think Gail Bukovich could see both rear wheels at the same time. In addition, Dr. Limpert said that a vehicle speeding around a curve would leave marks "in the beginning of the curve," although he did not specifically relate this phenomenon to plaintiff's accident.

After Limpert's deposition, defendants moved to strike his testimony, citing its inconsistencies with the factual record. Specifically, defendants pointed to the absence of skid marks on the curved portion of the roadway, the presence of skid marks before or a short distance into the curve, Hornfischer's testimony that he had applied the brakes well before reaching the curve, and Gail Bukovich's testimony that the rear wheels "unlocked" separately rather than together. The court granted defendants' motion in an order dated August 8, 1989.

Defendants promptly brought a motion for summary disposition pursuant to MCR 2.116(C)(10), arguing that, without Dr. Limpert's testimony, plaintiff had no expert testimony to offer at trial and, hence, no case. The court granted the motion on March 5, 1990. Plaintiff appeals.

The value of expert testimony is recognized under the Michigan Rules of Evidence:

If the court determines that recognized scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise. [MRE 702.]

In addition, MRE 703 provides:

The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to him at or before the hearing. The court may require that underlying facts or data essential to an opinion or inference be in evidence. [Emphasis supplied.]

See also, e.g., Kohn v Ford Motor Co, 151 Mich. App. 300, 314; 390 N.W.2d 709 (1986).

The requirements for the admission of expert testimony are:

(1) the witness must be an expert; (2) there must be facts which require an expert's interpretation or analysis; and (3) the witness' knowledge must be peculiar to experts rather than to lay persons. [ Jack Loeks Theatres, Inc v Kentwood, 189 Mich. App. 603, 611; 474 N.W.2d 140 (1991), citing O'Dowd v Linehan, 385 Mich. 491, 509-510; 189 N.W.2d 333 (1971).]

The decision whether to admit expert testimony is within the sound discretion of the trial court and will not be reversed on appeal absent a clear abuse of discretion. King v Taylor Chrysler-Plymouth, Inc, 184 Mich. App. 204, 214; 457 N.W.2d 42 (1990). See also Mach v General Motors Corp, 112 Mich. App. 158, 164-165; 315 N.W.2d 561 (1982) (alleged defect in bulldozer); Miller v Ochampaugh, 191 Mich. App. 48, 63; 477 N.W.2d 105 (1991) (dramshop liability). The trial court abuses its discretion in an evidentiary matter where its ruling has no basis in law or fact. Mulholland v DEC Int'l Corp, 432 Mich. 395, 411; 443 N.W.2d 340 (1989).

An expert witness need not rule out all alternative causes of the effect in question, but he must have an evidentiary basis for his own conclusions. Id. at 414. This Court has held that an expert's opinion was objectionable because it was based on assumptions that did not accord with the established facts. Thornhill v Detroit, 142 Mich. App. 656, 658; 369 N.W.2d 871 (1985). In Thornhill, a wrongful death action, it was critical to the plaintiff's expert's theory of the case for the decedent to have been "frothing at the mouth" for only a limited time before the defendants (an EMS crew) came on the scene. There was testimony, however, that the "frothing" had begun earlier. The trial court excluded the testimony of the plaintiff's expert, id. at 661, and this Court affirmed. "[T]he opinion was irrelevant to the issues at trial as framed by the facts." Id.

We are convinced that the Thornhill rationale is applicable to the case at hand. Dr. Limpert's explanation of the accident required the truck's brakes to have locked at a point well into the curve. It would also predict the presence of skid marks at the point where the lock-up began. Yet the only skid marks that were observed were either before the curve or a short distance into it. Those skid marks, indeed, were consistent with an alternative explanation, that the driver (Hornfischer) had entered the curve too fast and begun to skid then, when he applied the brakes.

Hornfischer's testimony is also inconsistent with Dr. Limpert's analysis. While plaintiff is correct that Hornfischer was uncertain about his distance from the curve when he began to brake, there is no doubt that he testified that he braked when he saw the "Curve Ahead" sign, which was obviously before he reached the curve. In addition, Hornfischer explicitly described the truck's response to his applying the brake as "shaking and shuttering" or "going crazy," neither of which terms describes the sensations associated with a brake lock-up.

Dr. Limpert's theory would also require both the rear wheels to have locked up together, because the braking system sent too much hydraulic pressure to them both. That theory would also predict that the two rear wheels would "unlock" together. Yet the only witness who was actually able to observe both rear wheels during the towing operation was Gail Bukovich (Bloomfield testified that he could see only the left rear wheel) and he was unequivocal that the wheels were freed up separately.

When questioned about these inconsistencies, Dr. Limpert was not able to reconcile them other than by disparaging the witnesses' powers of observation. Thornhill, then, would direct that Limpert's testimony be excluded. The trial court did not abuse its discretion when it struck the Limpert deposition, i.e., there was a basis both in law and fact for its decision. Without Limpert's testimony, plaintiff had no proof that the truck was defective when it left defendants' control and, thus, no basis for his claim. Summary disposition was proper under MCR 2.116(C)(10).

Affirmed. Costs to defendant.