A plaintiff injured when cleaning out a combine's unloader brought this products liability claim against the combine's manufacturer. Finding the unloader's danger open and obvious, the district court directed a verdict in favor of the manufacturer. We affirm.

* Garland Melton's arm was caught in the auger of a vertical unloader of a combine manufactured by Deere & Co. The function of the vertical unloader is to discharge grain from the grain tank. At times, the grain tank and unloading system must be cleaned out. For this purpose, a five-by-five-inch opening is located at the bottom of the vertical unloader, covered by a door. Immediately above the door is affixed a decal:

! WARNING

To avoid bodily injury from turning auger, stop engine and remove start key before opening clean out door.

Another warning decal, located over a toolbox about two feet from the cleanout door, included the following instruction:

2. Disengage and shut off all engine and/or motor power before servicing or unclogging machine.

Despite these warnings, Melton and his co-workers used a method to clean out the unloader that required the engine to be running. They would empty the loose grain through the cleanout door, engage the auger in order to release grain that had been caught up, and then scoop out by hand this additional grain and residue.

At the time of the accident, Garland Melton was cleaning out the vertical unloader by this method. Working with him were his brother, Richard Melton, and Roger Newsome. Garland placed his hand and arm through the cleanout door in order to rake out grain residue. The engine was still running. Richard meanwhile had been kicking down grain from the grain tank. Seeing Newsome return with an empty bucket and not checking to see if anyone was at the cleanout door, Richard climbed into the cab and engaged the auger. Garland's arm was severely injured, requiring amputation.

Deere claimed at trial that several other methods can be used to clean out the vertical unloader without running the engine with the cleanout door open, including scraping out residue with a tool, washing out the unloader with a hose, or replacing the door when engaging the auger to release grain that is caught up. Melton's witnesses, on the other hand, testified that it was necessary to run the engine and engage the auger in order to clean out the grain.

Other, similar accidents with the unloading auger have occurred. Three of Melton's witnesses, for example, were injured in similar accidents with Deere combines. The trial court, however, limited Melton to evidence of those three accidents, all of which predated Melton's accident, and excluded the testimony of six other witnesses. At the time Deere manufactured the combine that injured Melton, the company knew of four accidents at the auger site, and had begun affixing the warning decals. Some time later, Deere altered the design of the unloader. The new design incorporated a smaller cleanout door through which a hand does not fit.

Melton's complaint against Deere sought actual and punitive damages. The trial proceeded solely on the theory of strict liability under Mississippi law, with Melton alleging that the combine was defective and unreasonably dangerous and that Deere's conduct in failing to correct the known defect justified punitive damages. At the close of Melton's case-in-chief, the district court granted Deere's motion for directed verdict, holding that a reasonable jury could not find other than that the danger was open and obvious and that, as a result, the defendant was entitled to judgment as a matter of law. Melton appeals, arguing that the directed verdict was erroneous and that the issue of punitive damages also should have gone to the jury. Melton also contests two evidentiary rulings.

We first address the legal basis for the district court's directed verdict against Melton's claim. In announcing its decision, the district court first noted that Melton's case against Deere had proceeded solely on the theory of strict liability in tort. Mississippi has adopted the doctrine of strict liability as stated in Restatement (Second) of Torts, Sec. 402A (1965):

(1) One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property....

State Stove Manufacturing Co. v. Hodges, 189 So. 2d 113, 118 (Miss.1966). Specifically, Mississippi applies the "consumer expectations" test for when a product is unreasonably dangerous. Id. at 121; see also Restatement Sec. 402A, comments g and i. The district court stated that the danger associated with the cleanout door of the unloader was open and obvious and therefore could not give rise to liability, citing Gray v. Manitowoc Co., 771 F.2d 866 (5th Cir.1985). In Gray v. Manitowoc, we held that "the patent danger bar adopted by the Restatement was incorporated into Mississippi's doctrine of strict liability." 771 F.2d at 869. In other words, a product that has an open and obvious danger is not more dangerous than contemplated by the consumer, and hence cannot, under the consumer expectations test applied in Mississippi, be unreasonably dangerous. In Gray v. Manitowoc, this court reversed a judgment on a jury verdict for the plaintiffs where the danger had been open and obvious as a matter of law. 771 F.2d at 870.

Melton has suggested that Mississippi may now employ a test for unreasonable dangerousness other than that based on consumer expectations. In support of this contention Melton cites Whittley v. City of Meridian, 530 So. 2d 1341 (Miss.1988), a strict liability case in which the Supreme Court of Mississippi made the following statement:

In determining whether a product is unreasonably dangerous a reasonable person must conclude that the danger-in-fact, whether foreseeable or not, outweighs the utility of the product.

530 So.2d at 1347. Such a "risk-utility" test for unreasonable dangerousness is distinct from the "consumer expectations" test discussed above, and does not necessarily bar recovery when a danger is open and obvious. See W. Keeton, D. Dobbs, R. Keeton, and D. Owen, Prosser and Keeton on the Law of Torts Sec. 99 at 698-99 (5th ed. 1984). We cannot say, however, that Mississippi has altered its doctrine of strict liability. Whittley quoted section 402A of the Restatement as the law of Mississippi. 530 So. 2d at 1347. The sentence to which Melton points was merely a description following that quote. It was unaccompanied by any discussion of unreasonable dangerousness, the "consumer expectations" test, or the "risk-utility" test. Furthermore, that section of the opinion discussed a defense based on an intervening cause; the court did not actually apply the risk-utility test. Thus, we cannot conclude from this single sentence that Mississippi has adopted a new test for unreasonable dangerousness. Accordingly, consumer expectations are still the basis of Mississippi's test, and there is still no strict liability for a patent danger.

With this legal framework in mind, we review the facts and evidence of this case. On the motion for directed verdict, the district court was required to view all the evidence in a light and with all reasonable inferences most favorable to the nonmoving party, and grant the motion if no reasonable juror could arrive at a verdict in favor of the nonmoving party. Boeing Co. v. Shipman, 411 F.2d 365, 374 (5th Cir.1969). On appeal, we review the district court's decision on the motion for directed verdict de novo. Id. at 367, n. 1.

We agree with the district court that the testimony in this case left no question for the jury. Melton presented six witnesses, including himself, who were familiar with the combine that injured him. These witnesses testified that they knew the auger was dangerous when moving. Furthermore, the location of the auger just inside the cleanout door was unobscured and familiar to all the witnesses. They testified that they would not put a hand through the door if they knew the auger would be engaged. It may be true, as Melton and one other witness, Adams, testified, that a nonmoving auger is itself not obviously dangerous. Nevertheless, the risk that the auger may be engaged whenever the engine is running is a matter of common sense with which a reasonable user of the combine must be charged. Indeed, there is no question here that Melton knew that the lever for activating the auger was in the combine's cab, and that nothing on the machine or elsewhere prevented anyone from engaging the auger at any time that Melton's hand was in the unloader while the engine was running. Melton testified that he knew that the auger would be activated from time to time during the cleanout process he and the other workers were using, and that he knew he would be hurt if the auger were activated while his hand was in the cleanout opening. We find, therefore, that the only reasonable conclusion from this testimony is that the danger of injury from the auger while the combine's engine was running was open and obvious. As a result, Deere was entitled to a judgment against Melton's strict liability claim as a matter of law.

Melton relies on two particular items of evidence to support his argument to the contrary. First, he maintains that the unloader was unreasonably dangerous in that, once engaged, the auger started up and reached top speed faster than expected. Melton's brother, Richard, and their supervisor, Danny Campbell, both testified to this effect. Both of these witnesses also testified, however, that they had seen the auger start up before. There was no contention that the combine acted differently on the day of the accident than on any other day. In addition, Melton testified that he did not actually think he would have time to get his hand out of the door if the auger were engaged. Thus, the testimony of these two witnesses regarding the start-up speed of the auger is not substantial enough to allow a reasonable juror to conclude that the auger started up faster than a normal user would expect.

Melton also points to a second particular item of testimony as sufficient to defeat Deere's motion for directed verdict. Specifically, each witness stated his opinion that the combine's cleanout area was more dangerous than expected. This testimony, however, is unsupported by any other facts. Except for the allegation about the start-up speed of the auger, discussed above, the witnesses did not explain why they could not appreciate the full danger when they knew that a moving auger is dangerous and when the risk that the auger may be engaged whenever the engine is running is apparent. Without some explanation, this conclusory testimony is not substantial enough to raise a jury question. Accordingly, the directed verdict in favor of the manufacturer was proper.

Melton also presents two challenges to evidentiary rulings by the district court that, if meritorious, would change the evidence on the record and could, therefore, alter our review of the propriety of the directed verdict. First, Melton claims that the court improperly restricted the testimony of an expert witness, John Sevart. Melton sought to have Sevart, a mechanical engineer experienced with combines and other agricultural equipment, state his opinion whether the combine in question was defective or unreasonably dangerous. The district court excluded this proposed testimony.

"The admission or exclusion of expert testimony is a matter left to the discretion of the trial judge and his or her decision will not be disturbed on appeal unless it is manifestly erroneous." Smogor v. Enke, 874 F.2d 295, 297 (5th Cir.1989) (quoting Perkins v. Volkswagen of America, Inc., 596 F.2d 681, 682 (5th Cir.1979)). Melton argues that Sevart's testimony was relevant and admissible, and cites various cases where expert testimony about the dangerousness of a product was admitted even though the question of an unreasonably dangerous defect turned on consumer expectations. See, e.g., Lenoir v. C.O. Porter Machinery Co., 672 F.2d 1240 (5th Cir.1982). Melton's argument ignores our standard of review. As we put it in a recent case, "simply because it would not be error to admit certain expert testimony does not mean it is error to exclude that testimony." Smogor, 874 F.2d at 297. Indeed, the idea of discretion necessarily means that the court has room to decide the issue either way without committing error. The district court's ruling on Sevart's testimony was not even remotely, let alone manifestly, erroneous. As the court noted, the only import of such testimony would have related to some theory of negligent design, on which the case was not being tried, or for a theory of strict liability that, as discussed above, is not the law of Mississippi. As to consumer expectations, Sevart's opinion would not have aided the trier of fact. The exclusion of Sevart's opinion, therefore, is not grounds for reversal.

Melton also challenges the district court's exclusion of evidence of similar accidents occurring after the date of manufacture of the combine that injured Melton. The district court excluded the testimony of six witnesses who had been injured in accidents at the cleanout door of similar Deere combines. Melton also had a document listing nineteen other serious accidents.

At the outset, we again note that the district court has discretion to admit or exclude evidence, including evidence of similar accidents. See Ramos v. Liberty Mutual Ins. Co., 615 F.2d 334 (5th Cir.1980). The court's reasons for excluding evidence of accidents post-dating the manufacture of the Melton combine are not clear. If the reason was that evidence of the later accidents was not relevant as a matter of law, see Fed.R.Evid. 401, we disagree. In a strict liability case, the dangerousness of a product does not necessarily depend on the date of manufacture or on the manufacturer's knowledge of a defect as of that date. See Jackson v. Firestone Tire and Rubber Co., 788 F.2d 1070 (5th Cir.1984). Here, evidence of similar accidents does tend to show that the cleanout system was more dangerous than the users contemplated. A better reason for excluding the evidence, however, is provided by Fed.R.Evid. 403, which provides:

Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.

The district court did allow evidence of three similar accidents. Thus, this category of relevant evidence was before the jury. It was within the court's discretion to exclude evidence of additional accidents as unnecessarily cumulative, even though relevant. Furthermore, a parade of accident victims carries a risk of unfair prejudice, and could confuse the issues since the question is not simply danger itself but unreasonable dangerousness as measured by consumer expectations. Thus, the district court did not abuse its discretion in excluding evidence of the additional similar accidents.

In conclusion, we hold that based on the evidence presented in Melton's case-in-chief, the danger of the combine's loading system was open and obvious, so that a jury could not reasonably hold Deere strictly liable. Furthermore, Melton's challenges to the district court's evidentiary rulings present no grounds for reassessing this conclusion. Because we affirm the district court's directed verdict of no liability, we have no need to address the issue of punitive damages. The judgment of the district court is

AFFIRMED.

Through the exercise of our diversity jurisdiction, federal courts have interpreted Mississippi strict products liability law in a manner that is not justified either by Mississippi state court precedents or generally accepted strict products liability doctrine. These federal decisions lead the majority to uphold the district court's conclusion that the danger presented by the combine auger in which Garland Melton's arm was injured was open and obvious and that this finding warranted entry of a directed verdict. I disagree with the circuit precedents on which the majority relies and with the outcome in this case. I would remand for a new trial.

A. When is a product design defective in Mississippi?

In State Stove Manufacturing Co. v. Hodges, 189 So. 2d 113 (Miss.1966), cert. denied, 386 U.S. 912, 87 S. Ct. 860, 17 L. Ed. 2d 784 (1967), the Mississippi Supreme Court adopted Section 402A of the Restatement (Second) of Torts as the appropriate standard for evaluating manufacturer liability in a products liability action. Section 402A provides that "[o]ne who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property." Restatement (Second) of Torts Sec. 402A (1964). The primary issue in a Section 402A design defect case is whether the product that caused the injury is unreasonably dangerous. The Mississippi Supreme Court explained in State Stove.

Ordinarily the phrase "defective condition" means that the article has something wrong with it, that it did not function as expected. However, where the article was made as intended, and yet proves to be not reasonably safe, the phrase "defective condition" has no independent meaning. The issue is whether the product is "unreasonably dangerous" or not reasonably safe.

State Stove Mfg. Co., 189 So. 2d at 121 (citations omitted). The Mississippi Supreme Court has reaffirmed this approach in subsequent design defect cases. See, e.g., Toliver v. General Motors Corp., 482 So. 2d 213, 217-18 (Miss.1985); Dunson v. S.A. Allen, Inc., 355 So. 2d 77, 78 (Miss.1978).

The court below interpreted Mississippi precedents as directing application of a "consumer expectation" analysis in determining whether a product design is unreasonably dangerous. Under that analysis, the only relevant evidence is the subjective expectations of ordinary consumers of a product as to its functions. The court explained when it initially rejected the testimony of a mechanical engineer.

Unreasonably dangerous means that the product did not meet the reasonable expectations of the consumer as to safety. A defective product means a product that does not meet the reasonable expectations of an ordinary consumer as to the safety of the product.

Now who is better qualified to say what the ordinary expectation, the reasonable expectations of a consumer as to safety of a product is than the consumer, these farmers, and their opinions, they are the consumers themselves....

The majority appears to have adopted the same analysis, as have other federal courts. See, e.g., Gray v. Manitowoc Co., 771 F.2d 866, 869, 871 (5th Cir.1985); Hedgepeth v. Fruehauf Corp., 634 F. Supp. 93, 98 (S.D.Miss.1986).

It is true that Mississippi has adopted Section 402A, which is often described as encompassing a "consumer expectation" or "consumer contemplation" analysis.1 A review of recent Mississippi design defect cases, however, indicates that if Mississippi ever did apply a consumer expectation approach as set forth in the court below, it has now moved beyond that analysis. These recent cases demonstrate that in determining whether a product is unreasonably dangerous, the primary focus should be the adequacy of the product's design.2 A plaintiff will prevail in an action by demonstrating that the design of the product fell below the standard of design to which the manufacturer is expected to adhere. In Toliver v. General Motors Corp., 482 So. 2d 213, 214 (Miss.1985), the plaintiff claimed that the Chevrolet Vega was defective because the gas tank "was inadequately designed to withstand punctures in rear-end collisions." The Mississippi Supreme Court held that the plaintiff could proceed under a strict liability theory and explained the proof required.

Comment g. to Section 402A defines defective condition as "a condition not contemplated by the ultimate consumer, which will be unreasonably dangerous to him." In the context of fuel tank design, obviously the plaintiff contemplated that the automobile which he purchased had a fuel tank affixed to it, which could become dangerous under some circumstances. Therefore, in order to make out his prima facie case, he must show that the placement of the tank on the car that injured him was defective: that it fell below the standard of automotive design contemplated by the user, and, thus, became unreasonably dangerous to him.

Id. at 218; see also Brown v. Williams, 504 So. 2d 1188, 1192 (Miss.1987) (quoting Toliver ); cf. Hall v. Mississippi Chem. Express, Inc., 528 So. 2d 796, 799 (Miss.1988) ("The proper focus in a strict product liability case is upon the utility and safety of the product in view of its intended function...."). Although the Toliver court uses the phrase "contemplated by the user," it does not use the phrase in the sense suggested by the court below--that is, as limiting the evidence to uninformed consumers' subjective expectations. Rather, the court's focus is on whether a reasonable person informed of a product's design characteristics and performance would consider it unreasonably dangerous. Thus, the court noted that in proving that a particular design is defective, a "plaintiff may introduce evidence of industry standards, to show deviation therefrom, or an alternate design, to show the feasibility thereof." Toliver, 482 So. 2d at 218. Other Mississippi cases confirm the relevance of this evidence. See Hall, 528 So. 2d at 799; Rose v. Mercury Marine, 483 So. 2d 1351, 1352 (Miss.1986) (quoting Toliver ); Ford Motor Co. v. Broadway, 374 So. 2d 207, 210 (Miss.1979).

Cases such as Toliver demonstrate the Mississippi courts' focus on a number of factors in determining whether a product is unreasonably dangerous and thus defective. This approach is consistent with the development of strict products liability law in the vast majority of jurisdictions and the corresponding rejection of the limited consumer expectations focus advocated by the majority and the court below.3 Although no Mississippi decision lists all the relevant factors a jury may consider, it is clear that alternative designs and industry standards are relevant. It is reasonable to assume that Mississippi courts would also consider engineering standards and a product's compliance therewith, the number and type of injuries resulting from a particular design, and the cost of eliminating a hazard as relevant to a determination of whether the design is unreasonably dangerous. Certainly no Mississippi case has rejected the relevance of such evidence.

The majority misreads the case law when it concludes that "consumer expectations are still the basis of Mississippi's test." The Mississippi decisions discussed above establish that the ultimate fact question for the jury is whether the product is unreasonably dangerous. The plaintiff may present evidence of the design's danger, industry and engineering standards concerning the design, and alternative designs and their costs. Considering all of this evidence, the jury then decides the design's reasonableness.

B. What is the role of the open and obvious rule?

In Gray v. Manitowoc Co., 771 F.2d 866, 869-70 (5th Cir.1985), a panel of this court held that the open and obvious or patent danger rule serves as an absolute bar to recovery in Mississippi strict products liability actions. See also Hedgepeth, 634 F. Supp. at 97-98 (holding that Mississippi strict products liability law bars recovery for injuries caused by product hazards that are open and obvious). The court reached this conclusion notwithstanding the absence of any direct authority in Mississippi case law supporting application of the rule. The key to the Manitowoc decision was the court's determination that consumer expectations should be the primary focus in determining whether a product is unreasonably dangerous. See Manitowoc, 771 F.2d at 869. Under this analysis, a product design incorporating an open and obvious hazard could never be unreasonably dangerous, because it would never be more dangerous than an ordinary consumer would expect.

When the focus of the inquiry is shifted from consumer expectations to product design, however, it becomes clear that the open and obvious rule is not appropriately applied as an automatic bar to plaintiff recovery. The ultimate fact issue for the jury is the reasonableness of a product's design. A design with an open and obvious hazard may or may not be unreasonable. Only after a complete assessment of the full extent of the danger, in light of engineering standards, alternative designs, including their costs, and other relevant factors can a jury make a determination about reasonableness. Certainly, one would not conclude that because the potential harm of getting a hand caught in uncovered gears or in an unguarded press is open and obvious, the danger is therefore not unreasonable, even though the danger could be eliminated at a minimal cost by attaching safety devices.

Generally accepted strict products liability law recognizes that obvious risks may be unreasonable and that a manufacturer ought not to be able to escape liability by merely contending the danger was obvious to the consumer. "Liability may attach even though the danger was obvious where an unreasonable danger could have been eliminated without excessive cost or loss of product efficiency." 2 American Law of Products Liability 3d Sec. 28.70, at 82 (1987); see S. Speiser, C. Krause & A. Gans, 5 The American Law of Torts Sec. 18.80, at 896 (1988).

This does not necessarily mean that whether a hazard in a product design is obvious is irrelevant to the outcome of a case. The obviousness of a hazard may be available as a defense to bar a particular consumer from complaining. Indeed, there is some Mississippi case law suggesting that the assumption of risk defense might be available in a strict products liability action. See Alley v. Praschak Mach. Co., 366 So. 2d 661, 664-65 (Miss.1979); Nichols v. Western Auto Supply Co., 477 So. 2d 261, 263-64 (Miss.1985). If the defense is available, the obviousness of a hazard would be relevant in determining the appropriateness of recovery. A conclusion, however, that a plaintiff in a particular case should not recover because, for example, he assumed the risk of harm does not mean the product design is reasonable.

The district court in the present case directed a verdict for Deere & Company at the close of plaintiff's case because "there [was] not proof in [the] record of a latent hazard." The court based its holding on the two federal court decisions in Manitowoc and Hedgepeth. For the reasons discussed above, I believe these decisions misinterpret Mississippi law and thus improperly apply the open and obvious rule. That rule is inconsistent with an analysis that evaluates a number of factors concerning a product's design to determine whether it is unreasonably dangerous. No Mississippi decision has approved the holdings in Manitowoc or Hedgepeth or reached a similar result, and I do not feel bound in this diversity action to the errors of Manitowoc and Hedgepeth.

Even under the majority's approach, I believe the evidence is sufficient to create a jury question as to whether the hazard was open and obvious. As the majority notes, a nonmoving auger is not in itself obviously dangerous. The majority responds, however, that "the risk that the auger may be engaged whenever the engine is running is a matter of common sense with which a reasonable user of the combine must be charged." This approach is excessively rigid. As the Seventh Circuit has noted,

Whether a danger is open and obvious depends not just on what people can see with their eyes but also on what they know and believe about what they see. In particular, if people generally believe that there is a danger associated with the use of a product, but that there is a safe way to use it, any danger there may be in using the product in the way generally believed to be safe is not open and obvious.

Corbin v. Coleco Indus., 748 F.2d 411, 417-18 (7th Cir.1984) (applying Indiana law). In Banks v. Iron Hustler Corp., 59 Md.App. 408, 475 A.2d 1243 (1984), a Maryland appellate court made a related point.

Whether a particular danger is obvious or patent can depend on a number of things--the complexity of the machine, the knowledge, age, background, experience, intelligence, and training of the person injured, the extent to which his required contact with the device is routine and repetitive, whether he is subject to distractions, for example. It necessarily is a question of fact, then, and, if there is any dispute about it, the question is for the jury to decide.

Id. 475 A.2d at 1251; see also FMC Corp. v. Brown, 526 N.E.2d 719, 725 (Ind.Ct.App.1988) ("Even though a person may be aware of a general danger, defects in the product may create an additional danger which is not open and obvious.").

At trial, plaintiff put on five witnesses, beside Melton, who had varying degrees of farming experience and exposure to the type of combine that injured Melton. Three of the witnesses had been involved in accidents similar to Melton's. The other two witnesses had been present when Melton's arm was injured. Although the testimony of these witnesses varied in details, the substance was in large part consistent. They believed that to properly clean out the combine it was necessary to have the engine running and to occasionally engage the auger. They also testified that in cleaning out the combine it was necessary to use their hand to scrape excess grain out of the sump in which the auger is housed and that in undertaking this procedure it was necessary to stand in a position not within clear view of the combine cab and the mechanism by which the auger is engaged. Several of the witnesses testified that they had participated in the clean-out procedure a number of times.4 They testified that they were not aware of previous accidents involving John Deere Titan series combines, that they did not expect the auger to be activated while they had their hands in the sump, and that the combine was more dangerous than they expected it to be before the accidents. On cross-examination, the witnesses testified that they knew in general that augers are dangerous, that they knew they would be injured if their hand was in the auger when it was engaged, and that they would not have put their hand in the sump if they had known the auger was going to be engaged.

This evidence suggests that the witnesses believed there was a safe way to clean out the combine, and indeed, several of the witnesses indicated they had cleaned the combines using the process described numerous times without injury. A reasonable jury might well conclude that the combine's danger which injured Garland Melton was not open and obvious.5 The district court should not have directed a verdict.