Two issues are raised in the plaintiffs’ *398appeal from the directed verdict in their product liability action. The first is whether the trial court erred by ruling that plaintiffs’ expert, Sidney Beale, was not qualified to give an expert opinion as to the cause of mastitis in the plaintiffs’ herd of dairy cows. The second issue is whether the plaintiffs’ proofs were sufficient to survive the motion for a directed verdict, with or without Beale’s testimony as to the cause of mastitis in the Mulholland herd. We have concluded that both questions must be answered in the affirmative and that this matter must be remanded to the trial court for further proceedings.
i
FACTS
The plaintiffs, Robert and Betty Mulholland, operated a dairy farm in Montcalm County for many years. The Mulholland operation was small. About fifty to fifty-five cows were milked in an old barn. Feed was raised on 140 acres of the farm.
In 1979, the Mulhollands reached an agreement in which two of their sons, Richard and Randy, would return to the farm and the dairy operation would be enlarged to support the three families. According to the plan, the dairy herd would be expanded to two hundred cows by September of 1982. A new barn was also to be built, including a new milking parlor. Finally, a new milking system was to be installed in the barn.
The barn was completed and the new milking system installed in March, 1980. The equipment for the milking system was supplied by defendant DEC International Corporation and installed by defendant Tom Nelson Crawford, doing business as Tommy’s Refrigeration Service.
*399By the time the new barn and milking equipment were operational, the Mulholland herd was already up to seventy-one cows. Additional cows were added to the herd in April and October, 1980, as well as March, 1981. Most, if not all, of the additional cows were purchased from other farms.
Sometime after March, 1980, the Mulhollands observed that their cows were not "milking out.” The cows appeared to be uncomfortable and "steppy” or nervous around the milking parlor. On June 1, 1980, the Mulhollands went from two to three milkings each day in an attempt to get the cows to milk out. About that time, they noticed sores beginning to develop on the teat ends of the cows. Increased mastitis, an infection and inflammation of the udder, also was noticed. On July 21, the Mulhollands returned to two milkings each day, but sores continued to develop on the cows’ teat ends and mastitis continued to grow in the herd. The milk from infected cows could not be sold.
George Stuewer, a doctor of veterinary medicine, began working with the Mulholland herd after the new, expanded operation began. Dr. Stuewer treated the more acute cases of mastitis, while the Mulhollands themselves treated the milder cases. Dr. Stuewer noted the scabs on the teat ends of the cows, as well as the "pinched” look of the teat ends after the milking cups were removed. He was aware that this condition was associated with milking machine problems. While watching a milking, Dr. Stuewer also observed that the milk was "thrashing” in the machinery, rather than developing a smooth flow. Dr. Stuewer did not profess to be an expert on milking machines. He therefore suggested the Mulhollands consult with Sidney Beale. Dr. Stuewer had previously spoken *400with Beale and was impressed with Beale’s knowledge of milking machinery.
Sidney Beale was not a veterinarian, but did have a b.a. degree in agriculture with an emphasis in dairy science. Beale began working with milking machines in 1950 and had been employed as a consultant for about ten years. When he arrived at the Mulholland farm in November, 1980, Beale first observed a complete milking of the herd. He too noticed the sore teat ends, mastitis, the fact that the cows were not milking out, and that the milk was thrashing in the milk lines. Beale ruled out other problems in the milking parlor and then concluded that these problems were related to the configuration of the milking machinery. Specifically, Beale concluded that the use of a header in the vacuum lines was causing the thrashing in the lines, thereby pinching the teat ends and preventing the cows from being milked out. Beale did not examine other aspects of the Mulholland operation, but did prescribe changes in the configuration of the machinery which would allow for removal of the header.
The changes prescribed by Beale were implemented by Tommy’s Refrigeration Service in December, 1980. Afterwards, Beale, Dr. Stuewer, and the Mulhollands all noted a decrease in mastitis and an increase in milk production in the herd.
The Mulhollands filed suit in February, 1981, alleging a breach of warranty, negligence in the design and manufacture of the milking machinery, and a failure to warn. Extensive discovery followed, and trial commenced on April 16, 1985. In the course of Beale’s testimony, counsel for the plaintiff sought to establish a foundation for the witness’ testimony as to the cause of mastitis in the Mulholland herd. After a lengthy voir dire, the *401trial court ruled that Beale was not qualified to give such an opinion because he was not a licensed veterinarian.
No further proofs were offered by the plaintiffs and the defense counsel brought a motion for a directed verdict. Defense counsel argued that the plaintiffs had not established a prima facie case in that they had produced no evidence that the milking machinery had caused either an increase in mastitis or the consequential decrease in milk production. The trial court agreed and entered an order of directed verdict for the defendants.
The plaintiffs appealed the decision, arguing that the trial court erred in ruling that Beale lacked the necessary qualifications to express his opinion as to the cause of mastitis in the Mulholland herd. Alternatively, the plaintiffs argued that there was sufficient circumstantial evidence of the cause of mastitis to create an issue of fact for the jury. The Court of Appeals, in an unpublished opinion dated August 5, 1987, disagreed, affirming the trial court’s decision. We granted leave to appeal on both issues in an order dated March 22, 1988. 430 Mich 857 (1988).
ii
THE EXPERT TESTIMONY
A. THE QUALIFICATIONS OF SIDNEY BEALE
The Michigan Rules of Evidence provide:
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, train*402ing, or education, may testify thereto in the form of an opinion or otherwise. [MRE 702.][1]
It is a long-established principle of Michigan law that the qualification of an expert to render an opinion is a matter which rests in the discretion of the trial court. Appellate courts will interfere with a trial court’s ruling in this regard only to correct an abuse of discretion. People v Gambrell, 429 Mich 401, 407; 415 NW2d 202 (1987).
Wigmore notes and endorses the abuse of discretion review standard over trial court rulings as to the qualifications of experts. 2 Wigmore, Evidence (Chadbourn rev), § 561, pp 756-759. According to Wigmore, three considerations underlie the restrictive review standard in these matters. First, a determination of the expert’s qualifications in light of the proposed testimony often involves complicated factual reviews. Second, cross-examination of the expert is an "ample and sure safeguard.” Third, the matter of qualifications is considered to be too trifling to warrant appellate review.
The trial court’s ruling in this case touched upon a number of general concerns regarding the qualification of experts, but ultimately came to rest upon the matter of licensing. As the trial court explained:_
*403From the testimony here, from Dr. Stuewer’s testimony and from the other testimony, and I think from the Court’s knowledge of these areas from the law it can be said that the determination of the cause of mastitis is one which should require the basic expertise possessed by a licensed practitioner, in other words, a veterinarian.
* * *
This is a rambling discussion, but I want the record to be clear on my basis for the ruling. It is the Court’s view that in this circumstance, while Mr. Beale may possess great amounts of practical experience in this area, his answers here first of all satisfy me that he has not had the academic or the testing procedures that the Court feels are necessary for a person to hold himself out as an expert who can determine the cause of a medical condition in an animal.
We believe that the trial court abused its discretion in so ruling.
MRE 702 expressly provides that an expert may be qualified by virtue of his knowledge, skill, experience, training, or education. It does not refer to licensing as a method of qualification, much less as a requisite for the qualification of an expert. We do not believe that this omission was inadvertent. At best, a license is evidence of qualifications and thereby a useful shorthand in day-to-day commerce. Except in the most narrow legal sense,2 a license is not a qualification in itself. Even its *404value as evidence of qualifications is diminished in the courtroom where the expert is available and there is time for careful interrogation by both parties.
It is perhaps tempting to equate the word "expert” with the notion of a licensed professional. However, there is no basis for doing so in the text, of MRE 702. Even the generic word "expert” may be misleading, since this is not a qualification in itself, but merely the conclusion that the particular witness is qualified to give opinion testimony. As Wigmore explains:
[I]t is misleading to think of some witnesses as experts and others as non-experts. In a strict sense, every witness whosoever is an expert. In other words, the very fact that he is allowed to speak at all assumes that he is fitted to acquire knowledge on the subject; though in the vast majority of matters no demonstration of his fitness is needed. It is common and not unnatural to confine the term "experts” to witnesses whose fitness, by reason of the subject-matter, needs to be first shown. But while there is (as will be seen), a practical distinction between the instances in which the fitness must be expressly shown and the instances in which it need not be, that is no reason for ignoring the fundamental principle that every witness whosoever is and must be, by hypothesis, fitted or "expert” in the matter about which he is allowed to give his supposed knowledge.
In particular, it is a mistake to suppose that an "expert” must be a person professionally occupied upon the matter to be testified to. This is a mistake having its special origin in the doctrine of Opinion evidence, and can there better be considered (post, § 1923). It is sufficient here to note that the only requirement is that the witness must be *405fitted to acquire knowledge on the matter he speaks about; and,' if he is thus fitted, that it is entirely immaterial whether he acquired his fitness by being professionally concerned in such matters. [2 Wigmore, Evidence (3d ed), § 555, pp 634-635. Emphasis in original.][3]
As Wigmore further explains, licensing generally is not required for expert medical testimony, nor should it be required:
The common law, it may be added, does not require that the expert witness on a medical subject shall be a person duly licensed to practice medicine; but in some jurisdictions this requirement has been introduced by statute. Except as an indirect stimulus to obtain a license, such a rule is ill-advised, first, because the line between chemistry, biology, and medicine is too indefinite to admit of a practicable separation of topics and witnesses; and, second, because some of the most capable investigators have probably not needed or cared to obtain a license to practice medicine. [2 Wigmore (Chadbourn rev), § 569, pp 789-790. Citations deleted, emphasis in original.][4]
*406Michigan has long embraced the more practiced view of expert testimony. In Evans v People, 12 Mich 27, 37-38 (1863), it was explained:
Circumstances may make whole communities familiar with diseases not generally known elsewhere, and reasonably competent to manage ordinary cases of such diseases, and to recognize their symptoms. Such is often the case from necessity in new countries; and the same necessity leads to a more general knowledge of the extent to which a neighborhood has suffered from any prevailing sickness than is usual in populous towns. And it often happens that some persons having no general skill become very familiar with particular subjects.
It would be very unwise to exclude such evidence, merely because the range of the witness’s knowledge is limited. There are as many grades of knowledge and ignorance in the professions as out of them. The only safe rule in any of these cases is, to ascertain the extent of the witness’s qualifications, and, within their range, to permit him to speak. Cross-examination, and the testimony of others, will here, as in all other cases, furnish the best means of testing his value.
We continue to believe that the only safe rule is to ascertain with some specificity the range of the witness’ qualifications and to permit testimony within that range.
As we have outlined above, Sidney Beale has a b.a. degree in agriculture with an emphasis in dairy science and has been working with milking machines and mastitis since 1950. Beale’s education in this regard is informal. He explained that there was very little literature on mastitis "in the early years.” Beale testified that there is a lot of literature now, although some of it is factual and some is questionable. Despite what appears to be *407an inevitable lack of formal training in the area, Beale’s studies and experience are extensive.
From 1950 until 1955, Beale worked at Kraft Foods. Beale was assigned to quality control work with dairy farmers. The "big problem,” as Beale explained, was mastitis, and milking machines were thought to be involved in this problem. Beale consequently ran experiments with various milking machines in an attempt to reduce the incidence of mastitis in herds supplying milk to Kraft.
For the next twenty-five years, Beale worked for the Michigan Milk Producers Association. Nearly all of that time, Beale was involved in the production and marketing of milk. He supervised the laboratory and quality control program.
During ten of the last thirteen years with the mmpa, Beale also worked on the Milking Management Assistance program with a veterinarian from Michigan State University, Dr. Louis Newman. The program was designed to study the effects of milking machines on the incidence of mastitis in dairy herds. As Beale described the program, he worked with the milking machines from the layman’s side, while Dr. Newman worked with cows from the professional side. Together, they gained a lot of "input” as to the role of milking machines in the creation of mastitis in dairy herds.
In addition to this employment, Beale has conducted numerous seminars on milking machinery and mastitis. Indeed, it was at one such seminar that Dr. Newman informally proposed their research project. Beale also has published at least one paper in the area and was a regular columnist in Hoarde’s Dairymen. Beale is a member of the International Milk and Food Environmental Sanitarian Association and the National Mastitis Council.
Since 1978, Beale has been self-employed as a *408milking-management consultant. He has worked with farmers in fourteen states and two provinces of Canada. Beale estimated that he consults with about fifty farmers each year in that capacity. Over his total career, however, Beale has inspected milking operations on some 15,000 farms.
After carefully reviewing the record, we are satisfied that Beale was not simply a well-qualified expert, but perhaps the expert in this particular junction of science and industry.5 Indeed, we find it ironic that Dr. Stuewer’s testimony was cited in support of the trial court’s ruling, since it was this licensed veterinarian who suggested that the Mulhollands consult Beale. Certainly at that time, Dr. Stuewer was impressed with Beale’s knowledge in this area, and there is no indication that his opinion had changed by the time of the trial.
Beale was frank in his admission that he was not an expert on animal diseases in general. He was not, however, called upon for such testimony. Indeed, the fact that there was mastitis in the Mulholland herd was not disputed.6 Nor was Beale *409called upon to diagnose mastitis, although there was ample testimony in the record that mastitis is such a common disease of dairy herds that farmers themselves not only diagnose it, but also treat it with antibiotics. Beale was simply called upon to give an opinion as to the relationship between a particular milking machine defect and the most common of dairy herd diseases.7
We find a striking parallel between this case and Wolscheid v Thome, 76 Mich 265; 43 NW 12 (1889). In Wolscheid, the plaintiff brought suit under a theory of breach of contract for the failure of the defendant to provide adequate care for sheep entrusted to his care. Two witnesses, Eldridge and Caruss, testified as to the adequacy of the defendant’s care after examining the conditions at the defendant’s farm. The Court explained: *410Can it be said that Eldridge and Caruss might have given their opinion as to the adequacy of the conditions at the Thome farm, but not been allowed to give an opinion as to likely effects of inadequate conditions? We think not.8
*409It was proper for them to go there and examine the lay of the land, and the character of the soil, and as experts, if they were shown to be such,— and they undoubtedly were, being farmers and extensive growers of sheep,—to give their opinion as to whether or not it was a proper place to keep sheep in the winter-time; but it was not competent for them to state as to the condition of the sheds inside, unless it was shown, as it was not, that the condition was the same when the sheep were kept there, and then to state that, as they saw it, it was not a fit place to keep sheep in. [Wolscheid, supra, p 271.]_
*410Much has changed since the mid-nineteenth century. The proliferation of academic degrees and increasing specialization of labor is evident for all to see. Nevertheless, we believe that there is a world of wisdom apart from college campuses and urban high-rises. Now, as then, there is much to be gained from the practical student of the common arts. Beale exemplifies that wisdom. Beyond a doubt, his testimony would have assisted the trier of fact in deciding the cause of mastitis in the Mulholland herd.
*411We conclude that a review of the trial court’s discretion is warranted in this instance. First, the error of the trial court is one of law, not dependent upon the complex facts of record. Second, because the expert was not qualified by the trial court, there was no cross-examination, nor was there competing expert testimony to safeguard against error. Third, quite obviously, we believe that the error was of significant magnitude, both in the context of this case and Michigan jurisprudence in general, to warrant review. Having reviewed the record and having found no basis in law or fact for the trial court’s ruling, we can only conclude that it constituted an abuse of discretion.9
B. THE EVIDENTIARY FOUNDATION FOR BEALE’S TESTIMONY
In addition to qualification of the witness, there must be facts in evidence to support the opinion testimony of an expert. O’Dowd v Linehan, 385 Mich 491, 509-510; 189 NW2d 333 (1971). The defendants alternatively argue that, even if the trial court had qualified Beale as an expert as to the cause of mastitis in the Mulholland herd, it would nevertheless have had to exclude Beale’s testimony because it was based upon inadequate factual data.10
The Michigan Rules of Evidence provide in relevant part:_
*412The 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. [MRE 703.][11]
Here, the defendants argue that there was an insufficient factual basis for Beale’s opinion because neither Beale’s own perceptions nor facts made known to him at or before the hearing would allow Beale to rule out the other potential causes of mastitis in the Mulholland herd. Implicit in the defendants’ argument is the premise that ruling out the other potential causes was essential to Beale’s conclusion.
It is undisputed that the immediate "cause” of mastitis is a bacterial infection of the udder. On the other hand, as Beale testified, the intermediate "cause” may be many different things with which a cow’s teats come in contact. Beale explained that to know fully how any particular cow contracted mastitis, it would be necessary to eliminate a number of these potential causes. However, Beale further explained that a milking machine may mediately "cause” mastitis by damaging the teat ends, thereby making the cow more susceptible to bacterial infection upon contact. Plaintiffs’ theory clearly and properly rested upon this mediate "cause” of mastitis in the Mulholland herd.12
*413Beale’s own perceptions at the Mulholland farm provided an ample basis for the conclusion that a defective milking machine caused the mastitis in the plaintiffs’ herd by making the cows more susceptible to infection. As we have noted, Beale observed a complete milking of the Mulholland herd on his first visit to the farm. He noticed that a number of the cows had sore teat ends and mastitis. Beale also inspected the milking machinery in particular.13
It is, of course possible, as the defendants suggested throughout the trial, that the true or more immediate cause of the mastitis was improper bedding, unsanitary stalls, or even mud in the barnyard. Neither Beale’s own perceptions nor those made known to him at or before trial would allow this expert to rule out these possibilities. Nevertheless, we do not find the greater wisdom in a rule which would require an evidentiary basis of this sort. To the extent that they are credible, the absence of an evidentiary basis upon which an expert may rule out other potential causes may reduce the credibility of the expert.14 To the extent *414that other potential causes are substantiated by the evidence of record, they may also support a verdict of comparative negligence.15 However, to require for each expert an evidentiary basis sufficient to negate all of the possible causes which might be asserted by opposing counsel would virtually eliminate expert testimony.16 We require only expertise of experts, not omniscience. In our view, it is sufficient if the expert has an evidentiary basis for his own conclusions. See, generally, 7 Wigmore, Evidence (Chadbourn rev), § 1922, pp 26-29.
Here, the expert provided an ample basis in his own perceptions for his testimony as to the cause of mastitis in the Mulholland herd. We do not find a lack of evidentiary basis to be an adequate alternative ground upon which to uphold the ruling of the trial court. See Danielski v Lukomski, 204 Mich 304; 169 NW 887 (1918). Cf. Roberts v Young, 369 Mich 133, 137; 119 NW2d 627 (1963).
*415III
THE DIRECTED VERDICT MOTION
The trial court granted the defendants’ motion for a directed verdict, reasoning that the plaintiffs had failed to establish a causal relationship between the milking system and the increase in mastitis in their herd. It is, of course, well established in Michigan law that a prima facie case for product liability, under either a negligence or warranty theory, requires proof of a causal relationship between the defect and the damage of which the plaintiffs complain. Piercefield v Remington Arms Co, Inc, 375 Mich 85, 98-99; 133 NW2d 129 (1965); Caldwell v Fox, 394 Mich 401, 410; 231 NW2d 46 (1975); Kupkowski v Avis Ford, Inc, 395 Mich 155, 161; 235 NW2d 324 (1975); Holloway v General Motors Corp (On Rehearing), 403 Mich 614, 622; 271 NW2d 777 (1978). However, as we held in Holloway, p 622:
On a motion for directed verdict, the question is whether it is reasonable to infer from the evidence, direct or circumstantial, that the accident was probably caused by a defect attributable to the manufacturer. Questions of comparative probability are to be resolved by the trier of fact.
A plaintiff in a product liability action need not offer evidence which positively excludes every other possible cause. It is enough that the plaintiff establishes a logical sequence of cause and effect, notwithstanding the existence of other plausible theories, although other plausible theories may also have evidentiary support. Id., p 623. In reviewing the trial court’s ruling on a defendant’s motion for a directed verdict, we examine the testimony and all legitimate inferences that may *416be drawn in the light most favorable to the plaintiffs. Matras v Amoco Oil Co, 424 Mich 675, 681; 385 NW2d 586 (1986).
Here, the testimony of Sidney Beale would have established a logical sequence of cause and effect between the alleged defect in the dec system and the increase of mastitis in the Mulholland herd.17 We are aware of the defendants’ theories that the increase of mastitis was caused by the rapid introduction of new cows into the herd, the use of lime bedding, unsanitary stalls, and mud in the barnyard. Indeed, the plaintiffs’ proofs themselves provide some evidentiary basis for these theories. However, viewing the evidence, particularly the voir dire testimony of Beale, most favorably toward the plaintiffs, we can only conclude that a sufficient causal relationship between the dec system and the increase of mastitis would have been proven.18 We therefore conclude that the trial *417court further erred in directing a verdict for the defendants.19
IV
CONCLUSION
The judgments of the circuit court and the Court of Appeals are reversed, and the case is remanded to the circuit court for further proceedings. We do not retain jurisdiction.
Levin, Cavanagh, and Archer, JJ., concurred with Boyle, J.FRE 702 is essentially identical to MRE 702, except for the opening phrase "If the court determines that recognized . . . .” The text of FRE 702 provides:
If 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.
Thus, the Michigan rule emphasizes the role of the court in determining preliminary issues of admissibility in general, as well as the admissibility of scientific evidence under the Davis-Frye rule in particular. See People v Davis, 343 Mich 348; 72 NW2d 269 (1955) and Frye v United States, 54 App DC 46, 47; 293 F 1013 (1923).
The licensing of veterinarians is set forth in MCL 333.18802-333.18824; MSA 14.15(18802)-14.15(18824). MCL 333.18811; MSA 14.15(18811) does proscribe the diagnosis of animal diseases without a license in veterinary medicine. However, we are apprised of no statute or rule precluding evidence as to the cause of an animal disease from those not licensed as veterinarians. In any case, we are inclined to view the requirement of licensure for diagnosis under this statute as a limitation upon diagnosis for the purpose of treatment, rather than expert testimony. Moreover, it is clear from this record that the existence of mastitis in the Mulholland herd was established by the testimony of licensed veterinarian, Dr. Stuewer, and was, in *404any event, undisputed by the defense. See n 5. See also anno: Admissibility of opinion evidence of lay witnesses as to diseases and physical condition of animals, 49 ALR2d 932.
The same, practical concerns are dispositive under FRE 702. Thus, as one treatise explains:
Rule 702 recognizes that it is the actual qualifications of the witness that count, rather than his title, by providing that an expert can be qualified by "knowledge, skill, experience, training or education.” As was pointed out in ¶ 702[02], supra, the court will initially have to make a relevancy determination as to whether there is an issue in the case that can be resolved by this particular expert. It will then have to decide what the appropriate qualifications are given the facts of the case. Just as the wrong title may mean that the witness is nevertheless qualified, the right title will not suffice if the witness does not have the qualifications required by the facts of the case. [3. Weinstein, Evidence, ¶ 702(04), pp 702-51 to 702-52.]
Michigan has no statute prohibiting unlicensed professionals from testifying as experts. States which prohibit such testimony continue to constitute a small minority of the jurisdictions of this country. See 2 Wigmore, Evidence (Chadbourn rev), § 569, pp 789-790, ns 4-5.
According to Beale’s testimony, there were only two or three milking-management consultants in the United States.
That the diagnosis of mastitis in the Mulholland herd was not seriously disputed is indicated by the following colloquy between defense counsel and Beale:
Q. [Mr. Bremer]: Have you had any classes since you’ve been out in the specific cause of mastitis and how to determine what has caused it in a cow?
A. [Mr. Beale]: My training in that area would have been on a day-to-day basis with Dr. Lou Newman.
Q. He did not teach you how to be a veterinarian?
A. No, he did not.
Q. He did not teach you how to be a teat end lesion specialist within veterinary sciences?
A. Just how to recognize problems, is all.
Q. Well, it doesn’t take a lot of brains to see that a cow that squirts out lumpy stuff in her milk has got mastitis. Betty and Bob [Mulholland] can do that much, true?
A. True.
*409Q. It’s not hard to tell when they finally got the problem, it’s which of all these various things might have caused that. That’s the problem, isn’t it?
A. It’s not that big a problem.
Cf. Lewis v Bell, 109 Mich 189; 66 NW 1091 (1896), in which this Court held that a stable hand with two years of experience was not qualified to give an opinion as to the cause of death of a team of horses under his charge.
In People v Hawthorne, 293 Mich 15; 291 NW 205 (1940), Justice McAllister would have ruled that the testimony of an eminently qualified psychologist on the question of sanity was inadmissible on the ground that insanity is a disease and therefore falls within the study of medicine. A majority of the Court agreed with the justice’s holding, affirming the defendant’s conviction of manslaughter, but did so on the ground of harmless error. Justice McAllister’s views on the psychologist’s qualifications were thereby rendered obiter dictum. They nevertheless prompted an extensive and spirited discussion by the majority, who viewed the trial court’s ruling otherwise as error requiring reversal. The majority, in an opinion by Justice Butzel, concluded:
When a nonmedical is offered as an expert on subjects in the orbit of medical science, the trial court is put on guard and should take greater precaution in the preliminary inquiry to determine the witness’s qualifications and the extent of his knowledge than might be necessary when a graduate of a medical school is proposed. Yet it may well be that for some purposes, as where the issue concerns proper medical treatment, even a licensed physician would not possess sufficient knowledge in a particular branch of his calling to satisfy a trial judge who, within discretionary limits, insists upon a high standard of reliability. There is no magic in particular titles or degrees and, in our age of intense scientific specialization, we might deny ourselves the use of the best knowledge available by a rule that would immutably fix the educational qualifications to a particular degree. [Hawthorne, supra, pp 24-25.]
See, generally, Friendly, Indiscretion about discretion, 31 Emory L J 747 (1982). That there are limits upon the exercise of the trial court’s discretion in this matter has long been recognized in federal law. See Giannelli, Imwinkelried, Scientific Evidence, § 5-3, pp 154-156.
A trial court’s ruling which reaches the right result, although for the wrong reason, may be upheld on appeal. Washtenaw Co Health Dep’t v T & M Chevrolet, Inc, 406 Mich 518, 520; 280 NW2d 822 (1979). Fout v Dietz, 401 Mich 403; 258 NW2d 53 (1977).
MRE 705 further provides:
The expert may testify in terms of opinion or inference and give his reasons therefor without prior disclosure of the underlying facts or data, unless the court requires otherwise. The expert may in any event be required to disclose the underlying facts or data on cross-examination.
Here, the trial court apparently required prior disclosure of the underlying facts for Beale’s opinion, although that ruling is not evident in the record.
See, generally, Prosser & Keeton, Torts (5th ed), § 41, pp 263-272; *4133 Restatement Contracts, 2d, § 346, comment e, p 116 (both applying the "but-for” test for causation).
Beale also made a number of subsequent visits to the Mulholland farm, although it is unclear in the record the extent to which he observed the dairy operation in those visits.
As explained in 2 Wigmore, Evidence (Chadbourn rev), § 659, pp 897-898:
The third corollary of the general principle of knowledge (§ 656 supra) is that the witness’ knowledge (assuming it to be based on personal observation) must not appear to lack adequate data as its basis of inference. For example, when Sherlock Holmes tells his companion that the neatly dressed person who is seen passing on the street is a banker, the father of six children, and a German by birth, the ordinary intelligence wonders at a statement based on such apparently invisible data. Yet if a man passed by in working clothes daubed with lime and brickdust, the ordinary intelligence would admit that any observer had the right to assert positively that the man *414was a bricklayer. Thus it is that the law may reject testimony which appears to be founded on data so scanty that the witness’ alleged inferences from them may at once be pronounced absurd or extreme.
This principle, however, sound as it is in theory, can seldom have frequent application. When the source of knowledge is so insufficient, courts will rarely need to pronounce the formal exclusion of the testimony. Its weakness is self-exhibited. The risk of excluding a useful though small item of testimony is greater than the risk of admitting testimony capable of exaggeration. Cross-examination will usually furnish the exposure.
Michigan follows the rule of pure comparative negligence. Placek v Sterling Heights, 405 Mich 638; 275 NW2d 511 (1979). The rule has been applied in product liability actions regardless of the underlying theory for recovery. See Karl v Bryant Air Conditioning Co, 416 Mich 558, 569; 331 NW2d 456 (1982).
This is not to say that the preservation of expert testimony is an end in itself, but only that it would be irrational to exclude such evidence on the basis of the mere assumption that other asserted causes are either theoretically or statistically significant. But see n 17.
The defendants have, at times, argued that no actual damages were incurred, asserting that the milk production of the Mulholland herd actually increased after the introduction of the dec system. However, the support in the record for this conclusion remains unclear and this conclusion is contradicted by the direct testimony of Robert Mulholland. Moreover, the defendants apparently did not specifically state this issue as a basis for their motion, but instead relied upon the theory that no causal relationship was proven between the increase in mastitis and the dec system. We therefore decline to consider this issue on appeal. See MCR 2.515; 3 Martin, Dean & Webster, Michigan Court Rules Practice, Rule 2.515, comment 3, p 226.
This matter might be resolved differently if Beale had conceded, for example, that an improperly configured milking machine was one of ten possible, immediate causes of the mastitis, all with an equal probability of being the sole cause:
On the issue of the fact of causation, as on other issues essential to the cause of action for negligence, the plaintiff, in general, has the burden of proof. The plaintiff must introduce evidence which affords a reasonable basis for the conclusion that it is more likely than not that the conduct of the defendant was a cause in fact of the result. A mere possibility of such causation is not enough; and when the matter remains one of pure speculation or conjecture, or the probabilities are at *417best evenly balanced, it becomes the duty of the court to direct a verdict for the defendant. [Prosser & Keeton, supra, § 41, p 269.]
Beale made no such concession despite the suggestions of defense counsel.
Our resolution of this issue obviates the need to consider the defendants’ argument that causation may only be proven by direct, rather than circumstantial, evidence. We do note that the defendants’ assertion has generated essentially no support in this Court in the past. See, e.g., Holloway v General Motors Corp, 399 Mich 617; 250 NW2d 736 (1977), rev’d on reh 403 Mich 614; 271 NW2d 777 (1978).