delivered the opinion of the court.
This is an action brought by plaintiff, Horace G. Rockett, for personal injuries suffered when the Chevrolet Truck in which he was a passenger left the highway and overturned. Plaintiff claims that the 4-year-old vehicle had been defectively manufactured by defendant, General Motors, and that such defects were the proximate cause of the accident. At the close of plaintiffs case, the trial court granted defendant’s motion for a directed verdict and entered judgment accordingly. Plaintiff appeals.
Plaintiff contends that (1) the trial court erred in granting defendant’s motion for a directed verdict and (2) the cause should be reversed and remanded for a new trial with instructions that certain evidence offered at trial by plaintiff and denied by the court be admitted into evidence. The nature of the issues before us necessitates a detailed discussion of the evidence adduced at trial.
On May 22, 1966, plaintiff was severely injured while riding as a passenger in a 1962 Chevrolet Model C 1404 half-ton pickup truck. The injuries occurred when the truck left the highway and overturned on Route 66 outside of Dwight, Illinois. Plaintiff’s employer, South Suburban Safeway Lines, had purchased the subject truck from defendant on April 20, 1962. The purchase order called for the installation of heavy-duty rear springs. At the time of the accident, the truck was 4 years old and had been driven approximately 61,000 miles.
The circumstances of the accident were testified to by plaintiff, a State trooper who investigated the accident scene, and Roland Maerz, the driver of the vehicle. The accident occurred at about 1:40 p.m. on a clear, sunny, dry day; however, it was an extremely windy day with wind gusts up to 45 miles per hour. At the direction of their employer, Maerz and plaintiff were driving south from Chicago to Bloomington, Illinois, to perform road service on a company bus disabled by two flat tires. Maerz testified that the greatest speed he attained during their journey was 55 miles per hour, although plaintiff stated that at one point in time he noticed the speedometer read 60 miles per hour. They had driven on Route 66 for 5 to 10 miles prior to the accident. At speeds in excess of 45 miles per hour Maerz did not like the control of the vehicle and so he would decelerate. He testified that strong gusty winds buffeted the truck, especially when large trucks passed in the opposite direction. The passing of a truck caused him to go off course about 3 feet to one side.
Just prior to the accident, Maerz was driving in the right-hand lane approximately 1 foot from the pavement edge. He estimated the vehicle’s speed at no more that 45 miles per hour. At that location there was a 6-inch drop from the pavement to the highway shoulder. A large truck *219passed them in the opposite direction, and immediately, thereafter, both right wheels of their truck left the roadway and dropped over the 6-inch ledge to the shoulder. Plaintiffs vehicle travelled 86 feet in this fashion before regaining the pavement. The truck then skidded left across the roadway, overturned, and came to a rest upside down in the median strip. A bus tire, which was being hauled unsecured in the truck bed, flew or fell from the truck during the accident and was found in a ditch along the right-hand side of the highway. Upon inspection of the vehicle it was further learned that both right tires came partially off their rims. At the time of the accident the truck was carrying weight just 30 pounds short of the maximum capacity of that model equipped with heavy duty springs.
Paul Madsen, purchasing agent for South Suburban Safeway Lines, testified that he purchased the subject truck new from George R. Gibson Chevrolet in South Holland, Illinois. He did not know whether the rear springs were installed by the dealer. Nor did he know whether the truck had been equipped with heavy-duty rear springs as ordered. Previously, he had not requested George Gibson Chevrolet to install springs on any of the other company vehicles. Madsen further testified that to his knowledge the subject truck had never been in an accident. If it had, he would have known because he would have purchased the necessary parts. The only changes made to tire physical configuration of the truck were the installation of a hydraulic tailgate weighing between 400 and 500 pounds, and the placement of a sheet metal bottomer inside the truck box to facilitate the loading and unloading of salt. Madsen testified that he drove the truck in 1964, and that it steered smoothly like a pleasure car. To his knowledge the rear springs and shock absorbers had never been changed. Nor had a wheel alignment ever been performed. The truck had different tires than it had on the day of purchase, but they were of the same size. The company had. no available facilities to check or reset casters. Madsen saw the truck in Dwight, Illinois, on the day after the accident.
Plaintiff claims that the subject truck was defective in two respects at the time it left defendant manufacturer’s control and that the two defects, either independently or in concert, proximately caused his injury. He asserts that the vehicle was manufactured with grossly maladjusted front-wheel caster and that it had not been equipped with heavy-duty rear springs as ordered, but rather with much softer springs.
At the time of trial, the truck was not available to any of the parties. The record reveals that shortly after the accident, South Suburban Safeway Lines had it transported to Commercial Truck and Body Company where it remained undisturbed for approximately 3 months. Ultimately, *220the vehicle was cut apart for salvage purposes and the relevant parts are now buried under 15 feet of landfill. Accordingly, plaintiff produced no direct evidence of the actual caster-setting or of the rear springs with which the vehicle was equipped.
Plaintiff’s conclusion that the track had an improper caster setting was arrived at inferentially from Maerz’ recollection of the steering and riding characteristics of the truck. Maerz was the sole witness to testify about the driving characteristics of the track other than Paul Madsen who merely stated that it drove smoothly like a pleasure car. Maerz testified that he drove the truck shortly after its delivery and that its steering characteristics remained unchanged to the date of the accident. He observed that the truck was very sensitive to wind currents and that the passing of a large truck would cause it to move to one side or the other. Maerz testified that from his experience he found either no caster or a negative caster can cause this type of feel.
A test vehicle was constructed by plaintiff for experimental purposes from another 1962 model C 1404 Chevrolet half-ton truck and certain parts from a 1961 model C 1404 half-ton truck. Plaintiff contends that the model was an exact duplicate of the crash vehicle in all material respects except for its steering characteristics and the rear end suspension. Two of plaintiff’s expert witnesses, Dr. Uzgiris and Mr. SmidI, determined that negative caster would have to be set onto the front wheels in order to achieve the sensation orally described by Maerz.
Maerz further testified that on several occasions he had operated a hydraulic lift that had been installed on the rear of the truck. He observed that the rear of the track would drop 6 to 8 inches when using it to lift oil drams weighing from 385 to 410 pounds. Dr. Uzgiris interpreted this result to mean that heavy-duty springs could not have been installed in that vehicle because under the conditions described by Maerz, the rear springs would compress only about 2 inches. Dr. Uzgiris further testified that a soft suspension system would aggravate a problem caused by negative caster if a load were introduced into the bed of the truck.
Plaintiff contends that the trial court erred in directing a verdict for defendant. Consequently, we are concerned only with the narrow question of whether the evidence, when viewed in its aspect most favorable to plaintiff, so overwhelmingly favors defendant, that no contrary verdict based on that evidence could ever stand. Pedrick v. Peoria & Eastern R.R. Co., 37 Ill.2d 494.
In Suvada v. White Motor Co., 32 Ill.2d 612, our supreme court articulated the elements necessary to prove a prima facie case in strict liability actions. Plaintiff must prove (1) that the injury or damage resulted from a condition of the product; (2) that the condition was an *221unreasonably dangerous one; and (3) that the condition existed at the time it left the manufacturer’s control. We emphasize that no direct testi-. mony was introduced concerning the cause of the accident or the existence of the alleged defects. While it is true that a prima facie case may be established exclusively from circumstantial evidence, said evidence must have a reasonable probative force, for the jury will not be permitted to engage in mere speculation or imagination. (Shramek v. General Motors Corp., 69 Ill.App.2d 72, 216 N.E.2d 244.) Accordingly, unless it can be said that the evidence at bar gives rise to a reasonable inference that the injury resulted from a condition of the product, that the condition was an unreasonably dangerous one, and' that the condition existed at the time it left defendant’s control, then the case should have been removed from the jury’s consideration. See Larson v. Thomashow, 17 Ill.App.3d 208, 307 N.E.2d 707.
There is evidence in the record to suggest that a negative caster setting and soft rear suspension are in fact dangerous conditions, and consequently, a directed verdict cannot be justified for failure of proof in this regard. However, this is only one of three elements as articulated in Suvada. Assuming for the moment, but not admitting, that the evidence also supports a conclusion that the truck was defective at the time it left defendant’s control, plaintiff must still prove that the defect actually caused his injury. In other words, its must be a reasonable inference that Maerz lost control of the truck because the vehicle was not equipped with heavy-duty rear springs and/or had an improper caster setting. The evidence simply does not warrant such an inference.
The question of proximate cause is a question of fact the determination of which must have some reasonable basis in the evidence. This is not a case where the accident itself would suggest a defective product. Nor is this a case where a failure or malfunction of a product is clearly involved. As we stated the track was over 4 years old and had been driven in- excess of 61,000 miles at the time of the accident. Little was known of the actual use or misuse of the vehicle since the man who usually drove the truck died prior to trial. The accident occurred on an extremely windy day, so gusty in fact that the State trooper who investigated the accident testified that he experienced difficulty keeping his own car on the road. Maerz stated that at the time of tire accident he was driving 45 miles per hour and was hauling a payload just short of the maximum for a truck of that model equipped with heavy-duty springs. When the accident occurred he said it felt like a blowout and acknowledged that a sudden loss of air in one of the right tires could have caused the accident. In fact, Maerz testified that after the accident, he found both right tires were flat and partially off the rims. The cargo itself was *222not secured or restrained in any way, and one of the large mounted bus tires, weighing at least 235 pounds, was found on the side of the road where the right wheels initially left the pavement.
Thus, all of these factors, a 4-year-old truck that had been driven 61,000 miles, a heavy load, a shifting of that load, high wind, a 6-inch ledge from the shoulder to the pavement, and a sudden loss of air in the tires are all possible causes of the accident. We further note that Maerz returned the truck to the pavement without any significant reduction of speed. This too could have caused the accident as well as other possibilities of driver error.
It is true that plaintiff need not disprove all other possible causes. (Spotz v. Up-Right, Inc., 3 Ill.App.3d 1065, 280 N.E.2d 23.) It is equally true, however, that a jury may not engage in mere speculation and conjecture. (Shramek v. General Motors Corp., 69 Ill.App.2d 72, 216 N.E.2d 244.) In order to submit a case to the jury there must be a quantum of evidence sufficient to give rise to a reasonable inference that the asserted fact is true, for to arrive at that conclusion on a lesser factual basis necessarily would require sheer speculation. In the case at bar there exist numerous equally possible causes of the accident, and to conclude that those asserted by plaintiff were the actual causes could only be accomplished through an exercise of gross guesswork. Thus, on the basis of the causation element alone, the trial court acted properly in granting a directed verdict and removing tire case from the consideration of the jury.
We also believe that plaintiff failed to establish that the truck was defective at the time it left defendant’s control. No direct evidence was produced to substantiate this allegation. The only testimony concerning the actual condition of the truck on the day of the accident was that of a mechanic who salvaged certain parts of the truck. He stated that nothing appeared to be wrong with either the front or rear suspension.
Plaintiff would have the jury infer that the alleged steering and riding characteristics of the truck were caused by extreme negative caster and soft rear suspension, and then further infer that these characteristics were present at the time of manufacture. We note that prior to the accident, no claim ever was made to defendant of any defect in the truck. This is true even though the truck had been serviced on many occasions. Plaintiff’s conclusion that defendant maladjusted the front-wheel caster of the truck was arrived at inferentially from Maerz’ recollection of the feel of the truck 1 week after its purchase in 1962, and its feel 4 years later on the date of the accident. Maerz testified that in his opinion “either no caster or a negative caster alignment can cause this type of *223feel.” There was no evidence that the truck actually had a negative caster at the time of purchase.
Plaintiffs conclusion that the truck was equipped with soft rear springs was also arrived at inferentially from Maerz’ recollection of the feel of the truck which, in his opinion, indicated soft springs, and from his observation that the rear of the truck would drop 6 to 8 inches when the hydraulic lift was used to raise oil drums. This observation was used as the basis for an opinion by an automotive design expert that heavy-duty springs could not have been installed on the subject truck. Maerz never measured the drop of the truck, and the 6- to 8-inch figure is only his estimation. Furthermore, Dr. Uzgiris stated that a 6- to 8-inch drop of the rear of the truck does not constitute a similar deflection in the rear springs. Dr. Uzgiris admitted that the spring deflection would be substantially less, but was unable to calculate the actual deflection.
Plaintiff has piled inference upon inference; and logic would dictate that the more remote the inference, the more enfeebled is its probative force. We hold there is no evidence in the record, either direct or circumstantial, from which a jury reasonably could conclude that a condition of negative caster and/or soft springs existed at the time the truck left the control of defendant manufacturer. To so conclude would require sheer speculation and conjecture. While the granting of a directed verdict is indeed a delicate matter, precedent reveals that even stronger cases have been taken from the jury.
In Shramek v. General Motors Corp., 69 Ill.App.2d 72, 216 N.E.2d 244, a grant of summary judgment for defendant manufacturer was affirmed on appeal. Plaintiff claimed that the blowout of a left rear tire caused the car in which he was riding to leave the highway and overturn. Plaintiff admitted that he could not produce the tire and that he had no knowledge of its continued existence. The tire was never subjected to an examination which would establish the preexisting defect as the cause of the blowout. The court held that without any examination of the tire to elicit the cause of the blowout, plaintiff could not prove, either directly or inferentially, a case of negligence, breach of warranty or strict liability. The court stated that blowouts could be attributed to a myriad of causes, and the determination of why the tire blew out could be left only to speculation.
In Larson v. Thomashow, 17 Ill.App.3d 208, 307 N.E.2d 707, there was no question that the failure of a product caused an accident. The drive shaft fell out of a 28-month-old vehicle that had been driven only 24,000 miles. We held there that the mere.fact the drive shaft fell out was not sufficient to draw the inference that the product was defective when if *224left the control of the manufacturer. We upheld a directed verdict for defendant manufacturer.
■ In Briner v. General Motors Corp. (Ky. 1970), 461 S.W.2d 99, plaintiff brought suit against Universal Chevrolet and General Motors alleging that her injuries were caused by a mechanical defect in the steering system of her car that existed at the time of sale and that was improperly repaired by the leader. A directed verdict was entered in favor of Universal, and the jury returned a verdict for General Motors. There plaintiff brought the car to Universal complaining that “it was pulling and galloping and there was vibration in it.” Seven months later she wrote to General Motors and complained of vibration and lack of speed. Universal found and corrected a maladjustment of the ah pressure in the front tires. Universal further discovered that the air conditioning compressor was a cause of vibration and it was replaced. In August, 1964, road tests were conducted by Universal, but they were unable to find anything wrong. At the time of the accident in March, 1965, the car had been driven over 25,000 miles. At trial General Motors introduced evidence that there was no mechanical breakdown in the steering system. Plaintiff produced no direct evidence of a defective mechanical condition, but rather attempted to prove the defect solely through circumstantial evidence. She introduced the testimony of two expert witnesses who, in response to hypothetical questions, stated that abnormal vibration in an automobile, which could be caused by misalignment and other specified defects, could affect the steering mechanism. The accident occurred 7 months and 9000 miles after Universal had serviced the car. As in the instant case, plaintiff did not complain to the dealer about the improper operation of the vehicle. The court agreed with plaintiff’s contention that a casual relationship could be established by circumstantial evidence, but it held that the law requires that more than a possibility exist. The court stated:
“While reasonable inferences are permissible, a jmy verdict must be based on something other than speculation, supposition or surmise.”
The court concluded that “considering the length of time that elapsed after Universal’s last servicing of the automobile, the mileage driven, the uncertain character of the cause of the accident, we are too far out in the realm of conjecture to have a proven case of liability. There were just too many possible causes other than a mechanical defect attributable to Universal’s negligence.” The court went on to say that what was said about plaintiff’s claim against Universal applied equally, if not more so, to General Motors.
*225Plaintiff also urges that the failure- of General Motors to produce the accident vehicle which was avaffable to it gives rise to a presumption that an examination of the vehicle would produce evidence harmful to the defense. There is no merit to this claim. The vehicle was taken by South Suburban Safeway Lines, Inc., plaintiff’s employer, to Commercial Truck and Body Company where it remained and was subsequently destroyed as being a total economic loss. The defendant did not own, possess or control the track or any part of it.
Plaintiff also urges that the trial court abused its discretion in refusing to admit evidence of experiments conducted with a test vehicle. Apparently, the vehicle was employed by plaintiff’s experts for the dual purpose of reproducing the “steering feel” of the accident vehicle as oraUy described by Maerz and of observing the effect of loads on its heavy-duty springs. Plaintiff contends that the test vehicle was identical to the crash vehicle in all material respects, and that any dissimilarities go to the weight of the evidence and not to its admissibihty, citing Ramseyer v. General Motors Corp. (8th Cir. 1969), 417 F.2d 859. As stated in Ramseyer, perfect identity between experimental and actual conditions is not required. However, that case also recognizes a threshold issue well established in IUinois: Has there been a foundational' showing of substantial similarity between the tests conducted and actual conditions? Absent such a showing the evidence is incompetent.
We note from the outset that the decision of whether to admit or exclude evidence of experiments rests largely in the sound discretion of the trial judge, and his determination will not be overturned on appeal absent a clear showing of abuse of discretion. (Sherman v. City of Springfield, 77 Ill.App.2d 195, 222 N.E.2d 62.) In the instant case the record discloses numerous material dissimilarities between the test model and actual conditions, and accordingly, we can find no abuse of discretion by the trial judge in excluding the evidence. For example, the test model did not have a 450-pound hydraulic tailgate hanging from its extreme rear. It had an 8-cylinder engine while the truck had a 6-cylinder engine. It had new shock absorbers, rather than those of a truck that had been driven 61,000 miles in commercial service hauling. It had the same type of tires- as those on the accident vehicle, but of course, their mileage differed. No similarity to the steering system of the accident vehicle was shown. Finally, there were many inaccuracies in the reconstruction of road conditions as they existed on the day of the accident. The discrepancies were not slight and only tended to build up inferences upon inferences that there was a causal relationship between the aUeged defects and plaintiff’s injuries. Inasmuch as there was no showing of es*226sential similarity between the test conditions and the conditions as they existed at the time of the accident, we hold that the trial court did not err in refusing the evidence.
Other errors are alleged with regard to the court’s refusal of certain evidence. We have carefully examined the record and find that said evidence was properly excluded as being irrelevant. The trial court did not abuse its discretion.
In sum, plaintiff was unable to present prima facie proof that tire accident was caused by a dangerous condition of the truck that existed at the time it left the control of defendant. The trial court properly directed a verdict for the defendant.
We therefore affirm the judgment.
Judgment affirmed.
ADESKO, J., concurs.