ROGOSHESKE, JUSTICE.

In this death-by-wrongful-act action for damages arising out of a collision between two automobiles, defendants appeal from an order of the district court denying their motion to set aside a special jury verdict and grant a limited new trial on the issue of liability. The dispositive issues are (1) whether the trial court erred in admitting expert opinion testimony of an admittedly qualified engineer regarding the speed defendants' automobile was traveling at the time of the accident, and (2) whether it was error to deny a new trial upon defendants' claim, asserted for the first time in support of their post-trial motion, that the expert opinion was based on an erroneous scientific theory. Upon the record in this case, we find no reversible error in admitting such expert testimony and in denying defendants' motion for a new trial.

On the evening of May 17, 1969, plaintiff's husband, Napolean LeMieux, who had been out fishing during the day, was passing time at a resort in rural Rice County. A friend, Lawrence LaCanne, and LaCanne's wife and daughter appeared at the resort about 11:15 that evening. Both men consumed some alcoholic beverages there. Believing LaCanne intoxicate, LeMieux offered to drive LaCanne, his wife, and child home in their 1962 Ford automobile and to have a mutual friend, Gordon Bauer, follow them in order to return LeMieux to the resort. The accident occurred shortly after midnight when LeMieux, driving eastward on County Road No. 12, attempted to make a left turn into the LaCanne driveway. The LaCanne vehicle was then struck by a 1965 Chevrolet automobile traveling westward on County Road No. 12 and driven by defendant Steven Bishop, a 16-year-old high school student who had been issued his driver's license less than 2 weeks before the accident. Accompanying defendant were two 16-year-old companions. As a result of the collision, both LeMieux and LaCanne were killed and LaCanne's wife and daughter were injured. Defendant Steven Bishop and his two young friends were also injured.

The jury by special verdict found defendant driver 72-percent negligent and decedent driver 28-percent negligent and assessed plaintiff's damages at $41,500.

In determining the issue of liability and apportionment of fault, the jury was required to resolve two pivotal disputed fact issues, namely, plaintiff's claim of excessive speed of the Bishop automobile and defendants' claim that decedent driver was operating the LaCanne automobile when he was exceedingly intoxicated.

As the record reveals, apart from the 107-foot skid marks laid down by the Bishop vehicle, the tragic consequences of the collision, the extensive damage to the two automobiles, and a description by witness Bauer of how the LaCanne vehicle flew into the air when struck by the Bishop vehicle, no eyewitness opinion testimony as to the speed of the Bishop automobile was available to plaintiff. In contrast, defendant Steven Bishop testified that he was driving about 50 miles an hour when he first noticed the headlights of the LaCanne vehicle and up to the time he applied his brakes. Additionally, his front-seat passenger, Jessie Northrup, testified that he was not "going fast." Plaintiff therefore, in preparation for trial, quite understandably sought expert testimony of what is now sometimes referred to as an "accident reconstruction expert" or an "accidentologist." She retained Lloyd Brenna, a registered mechanical engineer, to investigate the collision and to testify at trial. Although the record does not show the precise date he was retained, he began his investigation on October 15, 1969, some 5 months after the accident. The trial began on November 9, 1970, and the record indicates that by April 22, 1970, defendants were notified that Brenna was investigating the accident and would testify at trial.

Brenna's opinion testimony was based upon the facts he observed in examining the damaged vehicles and the accident scene, including tests he performed to determine the drag factor or "coefficient of friction" of the highway surface, and the factual testimony of other witnesses. Such testimony is without substantial dispute. Rice County Deputy Sheriff Stanley Pacolt arrived at the scene shortly after the accident and immediately attended to the injured parties. Thereafter, he noted the precise position of each vehicle in reference to a particular landmark. The following morning, Deputy Pacolt returned to the accident site and measured by a "measuring wheel" the 107-foot skid marks caused by defendants' automobile as it skidded into the LaCanne vehicle. At trial, he diagrammed the accident, testified as to the measurements he had earlier made, and described the straightness and appearance of the skid marks laid down by the Bishop automobile. In addition, Brenna also considered the testimony of defendant Steven Bishop given at a pretrial deposition that the brakes on his car were in good working order.

Brenna testified that his independent investigation began on October 15, 1969, when he undertook a comprehensive inspection of the asphalt surface at the scene of the accident and of the two automobiles involved in the collision. At that time he took photographs of the accident site and the automobiles. In addition, on October 17, 1969, he took aerial photographs of the accident site and of the two vehicles. He completed his investigation and reconstruction of the collision on November 17, 1969, by using a test vehicle to conduct six test runs at the scene of the accident in order to lay down skid marks at varying speeds. The purpose of these tests was to aid Brenna in arriving at the proper "coefficient of friction" or "drag factor."

After fully detailing his investigative efforts, Brenna gave the following testimony, which was neither objected to on direct examination nor refuted on cross-examination:

"Q. What did you find for the coefficient of friction?

"A. .89 and, as I say, that doesn't have any — it is a dimensionless number, it is a number, it is a dimensionless number, .89."

Brenna was then asked a comprehensive hypothetical question requesting his opinion as to the speed of the Bishop vehicle immediately before defendant Steven Bishop applied its brakes. This hypothetical question was significantly modified after repeated objections were sustained by the trial court until the court was satisfied that it embraced substantially all of the relevant facts and testimony then in evidence which the jury could find to be true. In response, Brenna, by a process of synthesizing the directional components and the masses of the colliding vehicles, concluded that defendants' automobile was traveling at a minimum speed of 29 miles an hour at the time of impact. Brenna further testified that, utilizing the coefficient of friction of .89, "the skidmarks he [defendant driver] laid down gave me [Brenna] a value of 52 miles an hour." The two figures were then "simply added" by Brenna, who expressed his opinion that "just prior to application of [the] brakes," the minimum speed at which defendants' automobile was traveling was 81 miles an hour.

1-2. Once again we are called upon to decide whether the testimony of an expert witness concerning the speed at which an automobile travels was properly admitted. In our past decisions we have acknowledged that the speed of a vehicle is distinctly related to the physical sciences. Accordingly, we have consistently held that, although opinion testimony of eyewitnesses is preferred, where such testimony is nonexistent or of little aid to the jury in resolving the issue, a "witness qualified as an expert may give his opinion, based upon the length of skid marks, as to the speed of a motor vehicle involved in an accident." Grapentin v. Harvey, 262 Minn. 222, 226, 114 N.W.2d 578, 581 (1962).

Of course, it has long been recognized that the speed of an automobile lies in a field in which a lay person gifted with reasonable intelligence, given a fair opportunity to observe, and having ordinary experience with moving vehicles may give opinion testimony. Indeed, such opinions, "based on observation and experience, are often more reliable than a calculation of speed [by an engineering expert] based on unknown variables that can change the result materially if they are not precisely correct." Sanchez v. Waldrup, 271 Minn. 419, 426, 136 N.W.2d 61, 66 (1965). Moreover, it is common experience that lay eyewitnesses' opinions of speed are not only often more reliable but often more persuasive to a trier of fact than the opinions of hired and potentially biased experts.

Nevertheless, when viewed in the context of our adversary system, where eyewitness testimony as to speed is either nonexistent or unsatisfactory as to one or the other of the adversaries, it is proper to seek the aid of an expert. Assuming that the expert, after thorough investigation, is able to arrive at an opinion, the admissibility of that opinion will depend upon (1) whether there exist sufficient factual data to assure a reasonably complete and accurate reconstruction of the accident without indulging in speculation; and (2) more importantly, whether such opinion testimony will assist the triers of fact.

Underlying our rule governing admissibility of opinion evidence is the recognition that the speed of a vehicle can ordinarily be expressed only in the form of an opinion or conclusion and that nonexpert opinion testimony as to speed is foundationally based upon easily established facts, susceptible of specific contradiction and reasonably related to the perception of the lay witness, whose opinion will often be expressed in terms of indefinite miles per hour, as for example, "at least 60 miles an hour." A party having such a witness not only has the right but perhaps the duty to present this testimony. In contrast, expert opinion testimony as to speed based on skid marks is foundationally based for the most part on a partisan assumption of facts, unrelated to the witness' perception and often difficult of meaningful contradiction by cross-examination, with the opinion expressed in precise miles per hour. Such opinion testimony is not indispensable, and it should be offered and received only if the trial court determines it will aid the triers of fact. This is not to say that a qualified expert should be retained only in extraordinary motor vehicle accident cases but, rather, it is to caution litigants concerning such expert's secondary testimonial role and the necessity for the adversary to be prepared to challenge the expert's anticipated opinions by cross-examination or otherwise — all to the end that the triers of fact will be aided in determining the disputed fact issue with a clear understanding of the basis for his testimony.

3. In this case, as previously noted, the lay eyewitness testimony regarding speed was the testimony of the 16-year-old defendant driver and that of his companion, who merely declared that she did not believe defendant was driving too fast; the skid marks; and the testimony of Gordon Bauer, who followed the LaCanne vehicle and described how it flew into the air when struck by defendants' vehicle. We cannot say on this evidence that the jury was beyond the need of assistance from an admittedly qualified expert to resolve plaintiff's contention that defendant Steven Bishop was driving at a negligent speed before applying the brakes on his automobile.

Although defendants argue that the trial judge erred in admitting Brenna's opinion testimony because of an insufficient evidentiary foundation and because the testimony was given in response to an improper hypothetical question, our examination of the lengthy record satisfies us that the trial court's painstaking efforts did accomplish full prior disclosure of the underlying facts or data upon which Brenna's opinions were based, and that the hypothetical question he was asked neither assumed facts not in evidence nor failed to embody substantially all of the relevant facts admitted at trial.

Defendants, however, more vigorously contend that a new trial should have been granted because at the hearing on their post-trial motion it was conclusively demonstrated that Brenna improperly computed the data he had gathered by "applying erroneous derivations to a formula contrary to the law of physics." As we understand it, defendants assert that ascertaining the speed at which a motor vehicle travels immediately prior to application of the brakes cannot, in accordance with the law of physics, be done by adding the speed at impact with the speed necessary to establish the skid marks prior to impact. Rather, one must compute the square root of the sum of the squares of the two speeds.

At the outset, the testimony established that from an engineering standpoint it is not necessary to drive the automobile involved in an accident with the same load, same tires, and at the same speed it was traveling prior to collision in order to ascertain the proper coefficient of friction of the asphalt surface upon which the skid marks appear. It should also be borne in mind that Brenna did not simply assume a certain coefficient of friction from a general description of the road surface by using a chart or publication. The record indicates that he conducted his own tests not only at the scene of the accident but under conditions substantially similar to those that existed at the time of the collision as well. In short, Brenna, by carrying out his own independent experimentation, did precisely that which in the Sanchez and Grapentin cases this court implied an expert ought to do to obtain the most accurate coefficient of friction and thereby avoid a speculative or unreliable opinion.

It is true that Brenna "simply added" the factors in the formula noted. We, of course, need not pass on the question of whether speed can be accurately calculated by the procedure Brenna employed. We need only note here that the speed of the car before braking of 81 miles an hour which he estimated for the Bishop vehicle seems exceedingly high. Nevertheless, any error in calculations or in the assumption of facts or data upon which the opinion was based goes to the weight of the testimony, not to its admissibility. Cf. Smith v. Rekucki, 287 Minn. 149, 177 N.W.2d 410 (1970).

Brenna's testimony at trial that speed components could be added to ascertain the speed at which defendants' automobile was traveling prior to application of the brakes was undisputed and unchallenged. Even though, as previously indicated, the result of Brenna's computations seems high, the record does indicate that he verified by two other independent calculations his estimate of the speed at which the car was traveling before braking.

In concluding that the trial court did not err in denying defendants' motion for a new trial, we cannot ignore the fact that over 6 months prior to trial defendants were made aware that expert testimony would be offered by Brenna. It is true that defendants could not have effectuated any pretrial procedures aimed at discovering Brenna's written conclusions. Leininger v. Swadner, 279 Minn. 251, 156 N.W.2d 254 (1968). However, this prohibition extends "only as to such written 'conclusions' and not to the underlying facts which form the basis for the expert's conclusions." Leininger v. Swadner, 279 Minn. 257, 156 N.W.2d 259. No pretrial discovery efforts were directed at learning the underlying basis for Brenna's opinion.

Furthermore, defendants presumably chose not to retain an expert to challenge the credibility of Brenna's testimony but to limit their challenge to cross-examination and argument. Even though the scientific authority on which defendants base their claim that plaintiff's expert witness erred (Baker, Traffic Accident Investigator's Manual for Police [4 ed.]) was available prior to trial, it was not used in a manner that a scientific treatise may be used to confront an expert witness on cross-examination. Since there is no question the manual on which defendants rely was available before trial, the trial court properly could not regard it as "newly discovered" evidence under Rule 59.01(4), Rules of Civil Procedure.

Assuming arguendo that the traffic manual could have been admitted as competent evidence, the sole purpose for its introduction would have been to impeach or contradict Brenna's conclusion. In Albertson v. Albertson, 243 Minn. 212, 217, 67 N.W.2d 463, 467 (1954), however, we held that evidence "which is merely contradictory, impeaching, or cumulative cannot be made the basis of a new trial except under the most extraordinary circumstances." We are convinced, on the basis of the record before us, that such circumstances were not presented to the trial judge. In fact, the ultimate impact of defendants' contention lacks persuasive impeaching value since even they concede that, had the formula suggested by them been used by Brenna, it would have been determined that defendants' automobile was traveling at a speed of 59.5 miles an hour immediately before the brakes were applied.

We turn finally to defendants' most persuasive argument for a new trial, namely, that, assuming miscalculation of speed by Brenna, his testimony was so misleading and prejudicial as to deny defendants a fair trial. Our consideration of the record as a whole persuades us to reach the same conclusion implicit in the trial judge's denial of defendants' motion. Plaintiff's argument to the jury did not hold out Brenna's opinion testimony as the most significant evidentiary support for her claim of excessive speed. The jury would not have to find that the Bishop vehicle was traveling at 81 miles an hour before the brakes were applied and that it struck the LaCanne automobile at 29 miles an hour in order to find defendant driver negligent in exceeding the 50-mile speed limit. Indeed, if the calculation urged by defendants and supported by the accident manual were employed, the calculated speed of the Bishop vehicle, as we earlier noted, would still be approximately 10 miles over the lawful limit. Determination of speed is for the triers of fact. We cannot justify substituting our judgment on an issue so peculiarly within the jury's province when evidence other than the expert opinion testimony, such as the tragic consequences of the collision, the length and nature of the skid marks, and the eyewitness testimony describing the LaCanne vehicle flying into the air as a result of impact, would alone sustain a finding of excessive speed.

Additionally, considering the trial judge's careful precautions in determining the sufficiency of the foundation for the expert's opinion, his adherence to the rules governing its admissibility, and his comprehensive explanation of the basis for denying defendants' post-trial motion, we conclude that a new trial is not warranted.

Affirmed.