delivered the opinion of the Court.
This is an appeal from a judgment for a plaintiff in a products liability case. The sole question is whether the trial court erred in admitting into evidence a portion of a service bulletin issued by a manufacturer to its dealers in connection with a products-recall campaign.
In April 1979, Ernest C. Cutchin purchased a new 1979 model Harley-Davidson XLH 1000 Sportster motorcycle from a local dealer. A month later, he brought it back for brake repairs. He complained that the rear brake pedal was “hard to push down”, that the rear brakes were “squealing”, and that some of the hydraulic brake fluid had escaped. The dealer made certain repairs, but the problems persisted and Cutchin returned to the dealer for further brake service.
The dealer completed the repairs and delivered the motorcycle to Cutchin on August 31, 1979. Later that evening, Cutchin was riding along a residential street at a speed estimated at 20 to 30 m.p.h. Approaching a curve to the left, he attempted to apply the rear brake but discovered “there wasn’t anything.” As he drew closer to an automobile travelling ahead of him, Cutchin made another effort to apply the rear brake, but it failed again to engage. He downshifted the transmission from fourth to second gear and reached for the front brake lever. As he did so, the motorcycle “jumped to the right a little bit” and collided with a parked car. Cutchin sustained grievous injuries, one of which required amputation of his right leg below the knee.
Cutchin filed a motion for judgment against Gordon Harper, individually and t/a Gordon Harper Harley-Davidson Sales, Inc., *322and Gordon Harper Harley-Davidson (collectively, the dealer), and AMF, Inc., t/a Harley-Davidson Motor Co., Inc., the manufacturer. Cutchin alleged that his motorcycle “was negligently designed, constructed, manufactured and assembled” by AMF; that AMF and the dealer had breached express and implied warranties of fitness and merchantability; and that AMF and the dealer “knew, or . . . should have known that the motorcycle . . . was dangerous and defective” and failed to warn him of the danger. Cutchin further alleged that the dealer had failed to repair the brake system in a workmanlike manner as warranted.
At trial, Cutchin introduced Dr. James W. Moore, a professor of mechanical and aerospace engineering with industrial experience, as an expert witness. In his examination of Cutchin’s motorcycle, Moore “found that there was a binding” when he attempted to depress the rear brake pedal. Searching for the cause, he saw that the “stop bolt”, threaded into the end of the pedal shaft, was “rubbing on . . . the valve gear, cover . . . and it had cut and worn a groove in that cover.” When the bolt strikes the cover, he said, “ [i]t tends to stop the brake pedal from moving on down and tends to prevent application of the brakes.” Moore felt that the stop bolt was too long for its intended function.
According to the expert, the position of the stop bolt controls the amount of “free play” in the brake linkage leading to the hydraulic piston which actuates the brake. The position is adjusted by turning the threaded bolt in or out of the pedal shaft and securing the bolt with a “jamnut”. AMF’s production manual specified that the free play should be “a sixteenth of an inch at the [hydraulic] piston.” Moore measured the free play on Cutchin’s motorcycle at “three-sixteenths of an inch.” Such maladjustment of the stop bolt reduces the clearance between the hexagonal head of the bolt and the valve gear cover. The resulting excess in free play “delays application of the hydraulic in the sense that the brake pedal has to travel farther before it starts even moving the hydraulic piston.”
In light of the excess length of the stop bolt, its proximity to the cover, and the excess free play, Moore considered the design unreasonably dangerous and the defect one that could cause a brake loss such as that Cutchin described. In Moore’s opinion, once an engineer acquires knowledge, during the design stage or after production, that the stop bolt could strike the cover he “should take immediate action to rectify that error.”
*323AMF constructed for use at trial a mock-up of the rear brake system installed on 1979 models of a Sportster XLH 1000 motorcycle such as the one Cutchin purchased. Lance Onan, AMF’s products safety engineer, testified that the mock-up contained all the same parts installed in the 1979 models and that all components had been adjusted to meet all clearances and free play required by factory specifications. The mock-up was functional, and, in a demonstration before the jury, when the rear brake pedal was depressed, the stop bolt struck the valve gear cover and made a mark similar to that on the cover installed on Cutchin’s motorcycle.
Cutchin offered to introduce a copy of a “service bulletin” dated June 29, 1979, which AMF had issued to its dealers in connection with a products-recall campaign. Cutchin’s motorcycle was a 1979 model of the same make and contained the same parts as those individually identified by serial number in the bulletin, but the number of his motorcycle was not among those listed. At the trial court’s direction, Cutchin excised certain portions of the bulletin. Over the defendants’ objections on relevancy and public-policy grounds, the court admitted the revised bulletin as the plaintiffs exhibit. As revised, the bulletin contained the list of numbers and two paragraphs relevant to the issue on appeal:
A quantity of 635 XLH and XLS 1979 motorcycles were produced and shipped between March 30, 1979 and April 19, 1979 with a possible defect in the assembly of the brake pedal stop bolt jamnut. Jamnuts of affected motorcycles may not have been securely tightened at the time of factory assembly. This condition could allow the bolt to unscrew from the pedal shaft arm and allow the bolt head to contact the lower inside edge of the sprocket cover, possibly locking the rear brake in the applied position or prevent the rear brake pedal from being applied.
To adjust free play, loosen jamnut (2) and turn stop bolt (1) inward (clockwise) to decrease free play, or outward (counterclockwise) to increase free play. When 1/16 free play is obtained, tighten jamnut (2) against shaft lever (3) while holding stop bolt (1) in position. Recheck free play.
*324The court submitted the case to the jury under special interrogatories. The jury found in favor of the dealer on Cutchin’s repair claim. It returned a verdict awarding Cutchin $210,000 on his design and warranty claims against AMF and on his warranty claim against the dealer. Assigning numerous errors, the defendants appealed. Wé granted the appeal, limited to consideration of the error assigned to the evidentiary ruling.
AMF maintains that the service bulletin was irrelevant and unfairly prejudicial. Citing cases from other jurisdictions, AMF says that such evidence is admissible only when “the recall campaign include [s] the allegedly defective vehicle” and “concerns the same defect alleged by the plaintiff.” Cutchin’s motorcycle was not listed in the bulletin, the record shows that the jamnut on his stop bolt was not loose, and, AMF reasons, absent independent proof that the defect described in the bulletin caused the bolt so to bind against the cover as to cause the brake loss described by Cutchin, the bulletin should have been excluded. We disagree.
The thrust of the expert’s testimony was that the brake loss was caused by a stop bolt which was designed too long and secured by a jamnut in a position which allowed it to bind against the valve gear cover. Obviously, this was the basis of the jury’s finding that the manufacturer had breached its warranties. It is true that the primary concern reflected in the bulletin was AMF’s failure to tighten a jamnut during the production process. But the bulletin also addressed another concern. The dealer was instructed “[t]o adjust free play, loosen jamnut (2) and turn bolt (1) . . . [until] 1/16 free play is obtained”. This language tacitly recognizes the possibility that the stop bolt could be so positioned by a tight jamnut that free play would violate factory specifications and impair brake function.
This was precisely the kind of defect and effect described in the expert’s testimony and illustrated in the demonstration conducted with AMF’s mock-up. It was undisputed that the brake linkage in Cutchin’s motorcycle was the same in design and assembly as that in the mock-up which, in turn, was the same as that in the 1979 models of the vehicles referenced in the service bulletin.
If otherwise competent, any evidence “that tends to establish the probability or improbability of a fact in issue is admissible.” Horne v. Milgram, 226 Va. 133, 139, 306 S.E.2d 893, 896 (1983). We agree with the trial judge that the bulletin “was an admission . . . that the stop bolt head on this model could contact *325... the sprocket cover and prevent the rear brake pedal from being applied”, and we will uphold his ruling that it was relevant as corroborative evidence of proximate cause of the accident. We note, too, that the bulletin was uniquely relevant as proof of Cutchin’s allegation that both defendants had knowledge of the potential danger and violated their duty to warn.
The defendants also argue, as they did below, that the service bulletin should have been excluded on public-policy grounds. Specifically, they contend that admission of evidence of recall campaigns in a products liability suit may have the antisocial effect of deterring manufacturers from recalling products with dangerous defects for repair or replacement. “Recall evidence in this case should have been excluded,” they say, “for the same reasons that evidence of subsequent repairs are held inadmissible.”
A similar argument was advanced in Carey v. General Motors Corp., 377 Mass. 736, 745-46, 387 N.E.2d 583, 588 (1979). It was rejected there, and we reject it here. Recall campaigns are not voluntary; they are mandated, regulated, and enforced by federal law. 15 U.S.C. §§ 1411-20 (1982).
Finding no merit in the assignment of error, we will affirm the judgment.
Affirmed.