SIMON, Judge.

Plaintiffs, the survivors of Mr. Donny Lee Allison (Allison), brought this wrongful death action against Sverdrup Corporation (hereinafter Sverdrup), Sverdrup Parcel and Associates, Inc. (hereinafter "SPA"); Sverdrup Technologies, Inc. (hereinafter "STI"); and Dr. Virgil Flanigan. SPA and STI, formerly ARO, Inc., are both subsidiary corporations of Sverdrup. The case was tried in the Circuit Court of the City of St. Louis. At the close of plaintiffs' case, the trial court entered directed verdicts in favor of defendants Sverdrup, SPA, and Dr. Flanigan. Plaintiffs' claims against STI were submitted to the jury at the close of all evidence. The jury returned a verdict in favor of STI and against plaintiffs. Based upon the comparative fault instruction given, Instruction No. 9, the jury assessed Allison's fault at 100% and STI's fault at 0%.

Plaintiffs' second amended petition contained counts of strict liability for defective design, strict liability for failure to warn, and negligence against Sverdrup, SPA, STI, and Dr. Flanigan. Plaintiffs' claims against STI were ultimately submitted to the jury on a negligence theory.

Plaintiffs appeal the directed verdicts entered in favor of Sverdrup and SPA, and the jury verdict and judgment in favor of STI. The directed verdict in favor of Dr. Flanigan is not part of the instant appeal. Plaintiffs argue that the "trial court's granting of directed verdicts in favor of Sverdrup and [SPA] should be reversed, in that plaintiffs presented ample evidence from which the jury could have found that these defendants were liable to plaintiffs." Plaintiffs argue further that the trial court erred in refusing to instruct the jury on the rescue doctrine while instructing on the rule of comparative fault.

Defendants Sverdrup and SPA have both moved to dismiss plaintiffs' appeal with respect to the directed verdicts entered in their favor. Both motions relate the following: That on October 11, 1985, the trial court entered an order granting Sverdrup's and SPA's separate motions for directed verdict; that on January 31, 1986, plaintiffs filed notice of appeal in the trial court; that said notice of appeal did not specify any appeal from the October 11, 1985 orders; and that said notice of appeal indicated that the only order appealed from was the judgment on the jury's verdict entered in favor of STI, on October 28, 1985. Sverdrup and SPA argue that the only appeal taken by plaintiffs is from the judgment entered in favor of STI on October 28, 1985 and, consequently, the October 11, 1985 directed verdicts in favor of Sverdrup and SPA are not reviewable.

Rule 81.08(a) requires, inter alia, that a "notice of appeal shall specify ... the judgment or order appealed from." The plaintiffs concede that they failed to specify the order of October 11, 1985, granting Sverdrup's and SPA's motions for directed verdict, in their notice of appeal. Nevertheless, plaintiffs argue that this failure should not result in dismissal. Relying on Williams v. MFA Mutual Insurance Co., 660 S.W.2d 437 (Mo.App. 1983), plaintiffs contend that we should review the merits because defendants Sverdrup and SPA have not been prejudiced and were aware that plaintiffs intended to appeal from the directed verdicts, and because the issues are discernible from the briefs and record.

It is well settled that the formal averments contained in a notice of appeal are to be liberally construed in order to permit appellate review so long as the opposing party has not been misled to his irreparable harm. Weller v. Hays Truck Lines, 197 S.W.2d 657, 660 (Mo. banc), transferred by, 192 S.W.2d 677 (Mo.App. 1946). In an attempt to liberalize the judicial view of technical errors, the Weller court held "that a notice of appeal which can reasonably be construed as an attempt in good faith to appeal from a final judgment or appealable order shall be deemed sufficient." Id. at 660-61. Weller has been consistently followed. See, e.g., Williams, supra; Carson Union May Stern Co. v. Pennsylvania Railroad Co., 421 S.W.2d 540 (Mo.App. 1967); State ex rel. State Highway Commission v. Kendrick, 383 S.W.2d 740 (Mo. 1964); Triller v. Hellwege, 374 S.W.2d 104 (Mo. 1963). See also, the many cases cited at 2 Mo.Digest, Appeal Error, §§ 417-422.

We are aware of the cases cited to us by Sverdrup and SPA, holding that a fault in the averments of notice is ineffectual to bring an unspecified order before the court of appeals. See, e.g., Green Hills Production Credit Association v. R M Porter Farms, 716 S.W.2d 296 (Mo.App. 1986); Charles v. Ryan, 618 S.W.2d 220 (Mo.App. 1981); In re Marriage of E.A.W. 573 S.W.2d 689 (Mo.App. 1978); Donnell v. Vigus Quarries, Inc., 489 S.W.2d 223 (Mo.App. 1972); Brissette v. Brissette, 471 S.W.2d 691 (Mo.App. 1971). However, in none of those cases did the court find a good faith attempt on the appellant's part to appeal from the unspecified order and that the opposing party had not been misled to his detriment. See further, Pittman v. Reynolds, 679 S.W.2d 892, 893 (Mo.App. 1984); Anthony v. Morrow, 306 S.W.2d 581, 583 (Mo.App. 1957).

It is true that Weller and its progeny, except for Williams, supra, have dealt with a notice of appeal from an order overruling a motion for new trial, obviously an unappealable order, see Haywood v. Haywood, 527 S.W.2d 36, 37 (Mo.App. 1975), instead of from the underlying judgment and, therefore, the issues were discernible from the briefs and the record. That rationale, however, is no less persuasive here, as was shown in Williams, 660 S.W.2d 437.

Williams involved a personal injury action brought by a motorcycle passenger against her insurer and the driver's insurer. Summary judgments were granted in favor of both underwriters and the plaintiff appealed. On appeal the driver's insurer argued, among other things, that the plaintiff had failed to perfect her appeal by not specifying the exact judgment from which the appeal was taken. The plaintiffs' notice stated that the appeal was from "an appealable order and judgment." Although the court noted the technical rule violation, it declined to dismiss the appeal. The court stated that undoubtedly Rule 81.08 requires that appellants specify exactly what order or judgment is being appealed from. It found, however, that the driver's insurer had not been prejudiced by the plaintiff's failure to specify that the appeal was taken from the summary judgment in its favor. The insurer was deemed aware that an appeal had been taken from the summary judgment in its favor and, because there was no confusion between judgments or parties, the court decided to review the merits. The court noted that the rule violation did not hamper the review of the merits, presumably because the record was otherwise sufficient.

What was said in Williams, is controlling here. Both Sverdrup and SPA were aware or should have been aware that plaintiffs' appeal was taken from the trial court's directed verdicts in their favor. The issue clearly appears and was raised in plaintiffs' motion for new trial. Sverdrup and SPA were served with copies of plaintiffs' notice of appeal, on which they were listed as parties. Thereafter, they were served by plaintiffs with copies of the legal file, the transcript, and plaintiffs' brief in which plaintiffs argued that the directed verdicts in favor of Sverdrup and SPA were granted in error. Sverdrup and SPA have filed briefs in response, arguing that the directed verdicts were providently granted. Thus, we conclude that Sverdrup and SPA have neither been misled nor prejudiced. Furthermore, we view the actions of plaintiffs as a good faith effort to comply with the provisions of notice and we are able to easily discern, from the briefs and record, the issues raised. Sverdrup's and SPA's motions to dismiss are hereby denied.

The pertinent facts surrounding the death of Allison are as follows: Sometime during the early part of 1977, Dr. John Amos, a professor of engineering management at the University of Missouri-Rolla (UMR) campus, learned that the Coors Brewing Company (Coors) of Golden, Colorado had abandoned an experimental project designed to investigate the feasibility of converting waste material into natural gas. Coors was willing to donate the equipment used in their experiment to UMR, so long as the costs involved in transporting the equipment were borne by the University. Dr. Amos conceived of bringing the Coors equipment to an area near Pomona, Missouri, to use in the disposal of sawdust. However, nothing was done by UMR to obtain the equipment. It was doubtful whether UMR could obtain federal funding for this sort of project.

In the late fall of 1977, Dr. Amos contacted another engineering management professor at UMR, Dr. Yidirim Omurtag, and suggested Omurtag look into other possible uses for the Coors equipment. Dr. Omurtag worked on various extension projects at the University, but was primarily interested in alternative energy sources. Omurtag contacted Art Tschannen, the Energy Director at Coors, and made arrangements for an on-site inspection of the equipment at Golden, Colorado. Two professors from the University, Dr. Marshall Findley, professor of chemical engineering, and Dr. H.H. Sineath, professor of engineering management, performed the inspection.

Drs. Findley and Sineath returned from Colorado with a positive report. Based on the information collected by Findley and Sineath, Dr. Omurtag changed the concept for which the Coors equipment would be used. Rather than use it as a sawdust gasifier buried in the South Central Missouri Ozarks, Omurtag proposed to bring the equipment to UMR and use it in a biomass conversion research project, to take sawdust and other wood by-products and convert them into natural gas. Omurtag believed if the University used the equipment in research at the Rolla campus and paid the cost of obtaining the equipment, the U.S. Department of Energy could be persuaded to provide further funding for the project.

Dr. Omurtag's idea was not well received on the UMR campus initially. Dean James E. Halligan was skeptical about the idea and the interim Chancellor Dr. Jim C. Pogue was a professor of English, not a technical man, and so felt apprehensive about committing the University to the project. Omurtag also showed Dr. Virgil Flanigan, a professor of mechanical engineering, the complete files and technical reports on the Coors equipment in seeking support for his idea. Flanigan responded negatively, telling Omurtag that the Coors equipment was bastardized; not big enough to be commercial and not small enough to be experimental. Flanigan told Omurtag that he would not be willing or able to participate in the project because he did not believe in it. Dr. Omurtag, however, was not dissuaded and began to seek support from other quarters.

Omurtag contacted Leo Cram, the Extension Director at the University of Missouri-Columbia, and informed him about his idea and the potential for a big research grant from the federal government. Leo Cram directed Dr. Omurtag to Dr. Ardeth Emmons, the Vice President of Research for the University of Missouri System. Emmons found Omurtag's idea appealing and told him to contact Dr. Dave Fenton, who was at that time Vice President of Business Development at SPA. Emmons told Omurtag that if Fenton was of the opinion that the project was viable and supported it, he (Emmons) would find the money necessary to transport the Coors equipment to UMR. It happens that Emmons and Fenton were members of a St. Louis Civic organization called the Regional Commerce and Growth Association; specifically both sat on the Science and Engineering Committee. Fenton was also an active alumnus of UMR and was often willing to support projects that would enhance the reputation of his alma mater. Fenton agreed to look at the material concerning the Coors equipment as a favor to the University and Dr. Emmons.

Fenton thought Omurtag's idea looked promising and he asked Ron Williams, the Projects Manager at SPA, to become involved in the project because of his expertise in mechanical engineering. Fenton asked Williams to accompany a UMR professor of engineering to Golden, Colorado to inspect the equipment and to determine whether it was worth the expense of transporting it to Rolla. After Williams returned from Colorado, Fenton contacted Dr. Emmons and told him that Omurtag's idea should be pursued. Fenton testified that SPA had a long term interest in the Coors equipment and Omurtag's idea based on the potential for industrial application.

On or about March 15, 1978, Fenton and Williams came to UMR to discuss with Dean Halligan the funds required for obtaining the Coors equipment. Their efforts, along with those of Dr. Emmons, were successful in getting approval to transport the equipment to Rolla at the University's expense. Shortly thereafter, Dean Halligan told Dr. Omurtag and Dr. Virgil Flanigan, who had earlier refused to participate in the endeavor, to begin preparing a proposal for funding. Dr. Omurtag christened the project as "GROW," an acronym for Gasification Research on Wood. The project was to be developed as a biomass conversion system.

The proposal was drafted primarily during the summer of 1978. Dr. Omurtag was at the UMR campus and Dr. Flanigan was in St. Louis working at SPA for the summer. While at SPA, Flanigan continued to receive his regular salary from the University, in addition to the hourly wage he received from SPA. Dr. Flanigan testified that there were times he worked on the proposal at SPA, even though it was not part of his assigned duties there. Dave Fenton testified that he occasionally spoke with Dr. Flanigan, while Flanigan was at SPA, regarding the progress being made on the GROW project.

During the same summer, Dave Fenton and Ron Williams took Dr. Flanigan to the STI research facility in Tullahoma, Tennessee to meet Dr. Nathan Welch. At that time Dr. Welch was employed by STI. STI is a "high-tech" research center which had been involved in various aerospace and energy programs with the federal government. Dr. Welch had previously submitted a proposal to the Department of Energy for funding of a biomass conversion project similar to GROW. His proposal, however, had been rejected because the Department of Energy was not satisfied with the size of the unit, it had been too small.

Due to Dr. Welch's previous experience, Fenton suggested that Welch assist Flanigan and Omurtag with the process engineering they needed for the proposal. Welch prepared various flow diagrams and calculated flow rates, temperatures, and pressures necessary to write the proposal. Welch spent approximately four hours assembling the data. He also visited Flanigan and Omurtag at SPA's offices in St. Louis to discuss the proposal and later accompanied Omurtag and Flanigan on a trip to Washington, D.C. to present the proposal.

Dave Fenton, as Vice President for Business Development at SPA, tried to keep abreast of the progress being made on the proposal that summer. He kept in touch with Omurtag, Welch, and Flanigan with respect to the project's progress. Fenton testified that if the proposed technology turned out to be economically and technologically feasible it would be to the economic benefit of SPA to know about it. Omurtag made frequent visits to St. Louis to see Flanigan about the project at SPA.

Ultimately the GROW project proposal was submitted to the Department of Energy for review and funding. The project, as submitted, comprehended a three-phased gasification project. The project entailed a system to feed wood fuel, i.e. sawdust and other wood by-products, into a six foot long reactor (the equipment from Coors), from which a useful gas could be produced. In Phase I, the equipment was to be made operational and a low grade Btu gas was to be produced. (Btu is the abbreviation for British thermal unit — the quantity of heat required to raise the temperature of one avoirdupois pound of water one degree Fahrenheit at a specified temperature). The purpose of Phase I was to obtain "scaled-up data for low Btu gas production" in order to justify continued funding for Phase II. Phase II required planned modifications to the system in order to produce a middle grade Btu gas. Phase III involved further modification to obtain a high grade Btu gas that could be produced on a commercial scale.

Dr. Omurtag testified that prior to the submission of the proposal, a preliminary presentation had been made to Mr. Nello Del Gobbo of the Department of Energy. Drs. Flanigan and Welch, Dean Halligan and Dr. Omurtag were present at the meeting. The concept was explained to Mr. Del Gobbo and it was made to appear that UMR had a strong supporter and industrial partner. After the preliminary presentation, Mr. Del Gobbo recommended that a final proposal be submitted.

Once the GROW project was approved by the Department of Energy, STI entered into a consulting contract with the University of Missouri. By its terms Dr. Welch was to provide technical assistance which required his presence at the site. By August 1, 1979, the Coors' equipment was in place at the Rolla campus, and the University had been awarded an $800,000 contract for the project by the Department of Energy.

The machinery provided by Coors consisted primarily of a cylindrical reactor; a six foot cylinder of approximately 41 inches in diameter. The reactor was 21 feet high with removable "domed heads" and a platform located on the inside at the bottom that supported a "fluidized media bed." In Phase I, the reactor was heated to 1400 degrees Fahrenheit and a mixture of air and sawdust was introduced into the chamber. These elements would then break down into their component parts. "The idea being to get as much carbon monoxide, hydrogen, and low Btu hydrocarbons to produce a low Btu gas."

Phase I of the GROW project lasted approximately through the end of January, 1980. Throughout Phase I, Dr. Welch was present at the GROW project site at UMR. Although Dr. Welch was responsible for the ideas behind Phase II and III, once Phase I had been completed, Welch testified that he had no further involvement with the GROW project on site. Dr. Omurtag, who was the director of the project at that time, believed that STI was not fulfilling its obligations under the consulting contract and tried to suspend performance under the contract. Dr. Welch, however, contacted Dave Fenton and informed him that difficulties had arisen with Omurtag. Dr. Flanigan and Dr. Omurtag did not work well together. Welch knew Fenton was an alumni of UMR and wondered if Fenton could approach the new Dean, Dean Davis, and raise the issue with him. Fenton and Welch met with Dean Davis, informed him about the problems with Omurtag, and "suggested ... that something need[ed] to be done about it." Dr. Omurtag was replaced as project director by Dr. Flanigan shortly thereafter. Dr. Flanigan felt Welch's expertise was vital to Phase II of the project and they met once or twice to discuss the project between May and September of 1980.

During Phase I, the system used to feed the sawdust wood fuel into the reactor consisted of a 55 gallon drum set on top of the reactor. The sawdust was fed into the reactor by means of a "star valve." The sawdust would pass through the "star valve" and be deposited on a screw feeder and travel down into the reactor. The "star valve" was used to prevent the pressurized gases in the reactor from escaping back up into the unpressurized drum. It had been Dr. Flanigan's idea to use the drum feeder because of delays in the delivery of the feed system designed for the project. Dr. Welch testified that he had designed the "star feeder."

In June of 1980, the drum feed system was replaced by a new feed system for Phase II. The new system consisted of a thirteen foot by nine foot bin with a 20 inch manhole entryway at the top, covered by a hinged lid. The Phase II feed system did away with the need for a "star valve" because the entire feed bin was pressurized and, therefore, there was no need to seal off the outside environment from the pressurized reactor. The Phase II bin was mounted on steel legs, and the sawdust would flow at a controlled rate through various rotating screws, located at the bottom of the bin, into the reactor.

The concept for the Phase II feed system was provided by Galloway-Ratcliff, Inc. (Galloway) to Dr. Flanigan while he was still working at SPA. Dr. Welch had informed Flanigan the functions the feeder needed to perform, who in turn had relayed them to Galloway. After Flanigan returned to UMR, Galloway began sending him its ideas concerning the design of the system. Later, Galloway sent him a set of sample specifications that Flanigan "wrote up and submitted" in order to purchase the system.

In late September of 1980, the decedent, Allison, a UMR employee and a lab mechanic at the GROW project site, informed Dr. Flanigan that sawdust was building up on the interior walls of the feed bin and, that a device used to measure the level of sawdust inside the bin needed to be loosened and cleaned. The problem of having sawdust block up in the bin was not uncommon on the GROW project. Allison advised Flanigan that it might be necessary to go into the bin to clear off the sides and inspect the measuring device. Dr. Flanigan testified that he advised Allison not to go in without first "purging" the bin with fresh air for two to four hours to force out hazardous gases. Flanigan testified that he told Allison that anyone entering the feed bin should have a rope tied around him, a lookout on top, and a ladder. However, as of October 1, 1980, and prior thereto, the GROW project had no written safety procedures governing "purging," testing the bin environment for dangerous gases, or safe entry into the bin.

In the late summer of 1980, a student research assistant on the GROW project named George Loelkes had entered the bin to unblock sawdust that had collected therein. Loelkes used a ladder and a rope to enter the bin, but he did not "purge" it before doing so. He also wore a particle mask to keep from inhaling sawdust, but took it off in the bin to wipe perspiration from his face. He never received either written or oral instructions regarding tank entry safety procedures. Loelkes remained in the bin for 10 to 15 minutes without ill effect.

On September 30, 1980, Allison asked several student research assistants to help clean the feed bin the next morning and free the measuring device. Kurt Holekamp said that he would be available. On the morning of October 1, 1980, there were four people present at the GROW project site: Allison; Carl Brown, another UMR employee and lab mechanic; Kurt Holekamp; and Ahmet Omurtag, another student research assistant and the nephew of Dr. Yidirim Omurtag.

At approximately 8:30 a.m., Kurt Holekamp climbed to the top of the feed bin, followed by Ahmet Omurtag, to see what needed to be done to clean the interior walls of sawdust and free the measuring device. Holekamp had a long pipe which he used briefly to stir the sawdust. After realizing that the pipe was not going to accomplish the task, Holekamp and Omurtag discussed entering the bin. Holekamp explained that he had been in the bin before and decided to go in again. Holekamp lowered a knotted rope, attached to the top of the bin, inside the chamber and climbed down. When he reached the bottom, he knelt down and began digging in the sawdust for the jammed device. Less than a minute later, Holekamp attempted to stand, was unable to, and fell over unconscious.

When Omurtag saw what had happened, he called out to Allison who was in a nearby building. In response, Allison climbed to the top of the feed bin to see the situation. Omurtag asked Allison if he wanted him to get some gas masks he had seen hanging in a nearby shed. Allison responded that he did, and Omurtag retrieved several of the masks and returned to the top of the bin.

Omurtag cautioned Allison that entering the tank could be dangerous, but Allison replied, "we have to take the chance." Before descending into the bin, Allison called to Carl Brown and instructed him to call an ambulance. Allison then donned the gas mask and climbed down the rope. The mask was a canister type with no independent air supply; it was not self-contained.

Once he got to the bottom of the bin, Allison went over to check on Kurt Holekamp. Meanwhile, Carl Brown had climbed to the top of the feed bin to assist Omurtag. By that time, Allison had been in the bin approximately one or two minutes. Brown and Omurtag saw Allison trying to climb back up the rope and out of the bin, but before he could reach the top he lost his grasp and fell back to the bottom of the tank. At that point Brown went to call the Fire Department.

An ambulance arrived approximately ten minutes later. The bin was purged and Allison and Holekamp were removed from the tank. Both had died of asphyxiation from carbon monoxide poisoning. Allison was 44 years old when he died. He is survived by his wife, Jama, and their three children; Kimberlee, Don, and Cathy Sue.

In their first point, plaintiffs maintain that the trial court erred in refusing to instruct the jury on the rescue doctrine as applied to the actions of Allison while instructing on the rule of comparative fault. Plaintiffs argue that the rescue doctrine is recognized in Missouri and is clearly applicable to the conduct of Allison as supported by the evidence. Plaintiffs argue further that because the rescue doctrine was applicable, the trial court's failure to give appropriate instruction thereon constitutes reversible error.

The first instruction offered by plaintiffs and refused by the trial court was a verdict director dealing with STI's liability for placing Kurt Holekamp in a position of peril, and read as follows:

Your verdict must be for plaintiffs if you believe:

First, plaintiff Jama S. Allison was the wife of Donny Lee Allison and plaintiffs Kimberlee L. Allison, Don M. Allison and Cathy S. Allison were the natural children of Donny Lee Allison, and

Second, either:

defendant designed the feed system in such a way as to permit toxic gases from the reactor to go into the feed tank; or

defendant failed to provide adequate procedures and equipment for safe entry into the feed tank, and

Third, entry of personnel into the feed tank was reasonably foreseeable to defendant; and

Fourth, defendant, in any one or more of the respects submitted in paragraph Second, was thereby negligent, and

Fifth, as a direct result of such negligence Kurt Holekamp was placed in peril.

The second instruction, which was also offered and refused, concerned the concept of negligence within the context of comparative fault under the rescue doctrine. It read:

It is not negligence to knowingly and voluntarily place one's self in a position where the likelihood of receiving serious physical injury is great when the exposure to such danger is conducted in a reasonable manner and for the purpose of saving human life.

The trial court charged the jury as follows:

INSTRUCTION NO. 6

Your verdict must be for plaintiffs if you believe:

First, plaintiff Jama S. Allison was the wife of Donny Lee Allison and plaintiffs Kimberlee L. Allison, Don M. Allison and Cathy S. Allison were the natural children of Donny Lee Allison, and

Second, either:

defendant designed the feed system in such a way as to permit toxic gases from the reactor to go into the feed tank; or

defendant failed to provide adequate procedures and equipment for safe entry into the feed tank, and

Third, entry of personnel into the feed tank was reasonably foreseeable to defendant, and

Fourth, defendant, in any one or more of the respects submitted in paragraph Second, was thereby negligent, and

Fifth, as a direct result of such negligence Donny Lee Allison died.

The term "negligent" or "negligence" as used in this instruction means the failure to use the reasonable technical skill, ability, and competence ordinarily used under the same or similar circumstances by members of defendant's profession.

INSTRUCTION NO. 7

You must assess a percentage of fault to plaintiffs if you believe:

First, either:

Donny Allison directed Kurt Holekamp to enter the feed bin contrary to instructions, or

Donny Allison failed to properly supervise Kurt Holekamp when Kurt Holekamp was working on the feed bin, or

Donny Allison failed to purge the feed bin with air before Kurt Holekamp entered the feed bin, or

Donny Allison failed to follow feed bin entry instructions when he entered feed bin.

Second, Donny Allison in any one or more of the respects submitted in Paragraph First, was thereby negligent, and

Third, such negligence of Donny Allison directly caused or directly contributed to cause any damage plaintiffs may have sustained.

The term "negligent" or "negligence" as used in this instruction means the failure to use that degree of care that an ordinarily careful and prudent person would use under the same or similar circumstances.

Plaintiffs' first point raises initially the question, not previously considered by Missouri courts, as to the application of the rescue doctrine in a comparative fault situation. The rescue doctrine has long been recognized in Missouri. It was employed in an embryonic form in Donahoe v. The Wabash, St. Louis Pacific Railway Co., 83 Mo. 560 (1884) and its salutary purposes have retained their vitality through the present day. See Lowrey v. Horvath, 689 S.W.2d 625 (Mo. banc 1985). See also United States v. Pesano, 304 F.2d 856, 860 (8th Cir. 1962) (opinion by Blackmun, J.) and cases cited therein. The most oft cited articulation of the underlying rationale of the rescue doctrine comes from Justice Cardozo:

Danger invites rescue. The cry of distress is the summons to relief. The law does not ignore these reactions of the mind in tracing conduct to its consequences. It recognizes them as normal. It places their effects within the range of the natural and probable. The wrong that imperils life is a wrong to the imperiled victim; it is a wrong also to his rescuer. The state that leaves an opening in a bridge is liable to the child that falls into the stream, but liable also to the parent who plunges to its aid.... The risk of rescue, if only it be not wanton, is born of the occasion. The emergency begets the man. The wrongdoer may not have foreseen the coming of the deliverer. He is accountable as if he had.

Wagner v. International Railway Co., 232 N.Y. 176, 180, 133 N.E. 437, 437-38 (Mo. 1921) (citations omitted).

In Lowrey v. Horvath, 689 S.W.2d 625 (Mo. banc 1985), our Supreme Court stated:

The rescue doctrine offers two important benefits to the person who is attempting to recover for injuries incurred during a rescue attempt. First, under the rescue doctrine it is not negligence to knowingly and voluntarily place one's self in a position where the likelihood of receiving serious physical injury is great when the exposure to such danger is conducted in a reasonable manner and for the purpose of saving human life. McConnell v. PicWalsh Freight Co., 432 S.W.2d 292 (Mo. 1968); Hammonds v. Haven, 280 S.W.2d 814 (Mo. 1955). Second, the plaintiff may invoke the rescue doctrine to establish "that the defendant's negligence in creating the peril which induced the injured person to attempt to rescue another who was imperiled was the proximate cause of the injury for which recovery is sought * * *." 57 Am.Jur.2d Negligence § 418 (1971). See, e.g., Stone's Independent Oil Distributors v. Bailey, 122 Ga. App. 294, 176 S.E.2d 613 (Ga.Ct.App. 1970).

Id. at 626-27.

Prior to the decision in Gustafson v. Benda, 661 S.W.2d 11 (Mo. banc 1983) and the adoption of a comprehensive system of comparative fault under the guidance of the Uniform Comparative Fault Act; the rescue doctrine, simply stated, held that when a person is exposed to peril of life or limb by the negligence of another, the latter will be liable in damages for injuries received by a third person in a "reasonable effort to rescue" the one so imperiled. National Dairy Products Corp. v. Freschi, 393 S.W.2d 48, 57 (Mo.App. 1965); Restatement (Second) of Torts § 472 (1965); W. Prosser W. Keeton, The Law of Torts § 44, at 307-08 (5th ed. 1984). A "reasonable effort to rescue" included everything short of rash or reckless conduct. If the rescuer's conduct crossed the lines of rash or recklessness, he or she was precluded, as a matter of law, from recovering from the person whose negligence created the peril. Lowrey, 689 S.W.2d at 625.

The general rule applicable to a rescue case prior to the adoption of comparative fault is stated in Hammonds v. Haven, 280 S.W.2d 814, 816 (Mo. 1955) (quoting C.J.S., Negligence § 124):

Under what is commonly referred to as the rescue doctrine, conduct which might otherwise be considered contributory negligence may not be so considered where a person is injured in attempting to save others from imminent danger of personal injury or death. Persons are held justified in assuming greater risks in the protection of human life where they would not be under other circumstances. One is not guilty of contributory negligence in exposing himself to danger of injury in order to rescue another from imminent danger of personal injury or death if, under the same or similar circumstances, an ordinarily prudent person might so expose himself, or, as often expressed, if the act of intervention is not performed under such circumstances as would make it rash or reckless in the judgment of ordinarily prudent persons. This is true even though the person attempting to rescue knows that it involves a great hazard to himself without certainty of accomplishing the attempted rescue and even though in attempting such rescue he thereby imperils his own life.

Id.

Before comparative fault became the law in Missouri, the rescue doctrine served the important function of, inter alia, mitigating the harsh rule of contributory negligence which served to completely bar recovery if a plaintiff was guilty thereof. See Lowrey, 689 S.W.2d at 626-27. Because the law places a premium on human life, the rescuer was justified in exposing himself to a danger and in a manner that under other circumstances would deprive him of legal redress for injuries sustained. Doran v. Kansas City, 237 S.W.2d 907, 912 (Mo.App. 1951). As noted, so long as the rescuer's conduct was not rash or reckless he was entitled to recover. The rescue doctrine served to allow the rescuer to maintain an action, when he otherwise would have been barred by the doctrine of contributory negligence, so long as the rescuer's action did not rise to the level of rashness or recklessness. See, 57 Am. Jur.2d, Negligence, § 418 at 836 (1971); Lowrey, supra at 626.

Prior to Gustafson and the adoption of comparative fault, there were two ways in which a rescuer's conduct could preclude his recovery. First, the rescuer may have been rash or reckless in attempting the rescue, or in his method of carrying it out. If so, he was precluded from recovering from a defendant who had caused the rescuee's peril. Welch v. Hesston Corp., 540 S.W.2d 127, 129 (Mo.App. 1976). Second, the rescuer's negligence may have contributed to creating the very situation that later "invited" his rescue. In such a case, the rescuer's negligent conduct was deemed a contributory factor in producing his or her injuries and, as such, served as a complete bar to recovery against a defendant whose negligence also contributed to the third person's peril. See Dulley v. Berkley, 304 S.W.2d 878, 883 (Mo. 1957). See also, McConnell v. Pic-Walsh Freight Co., 432 S.W.2d 292, 299-300 (Mo. 1968).

It was only when a rescuer attempted a rescue that the rescue doctrine served to mitigate the astringent rule of contributory negligence. In such a case, the rescuer was entitled to recover unless he had acted rash or reckless. Lowrey, 689 S.W.2d at 626. However, if the rescuer had negligently contributed to the creation of peril that later "invited" his rescue attempt, his negligence was sufficient to bar recovery. See Dulley v. Berkley, 304 S.W.2d 878, 883 (Mo. 1957) (quoting the First Restatement Torts § 472 for the proposition that if the prior negligence of the plaintiff was a contributing cause of the peril to the person rescued, the plaintiff cannot recover for harm sustained in the attempt at rescue).

It is apparent from the foregoing that a rescuer is subject to two different standards of conduct depending on his actions. The standard of ordinary negligence governed if the rescuer had created the situation of peril inviting his rescue attempt. However, if another was in danger, then the rescuer could use all means to effect a rescue, short of rash or reckless.

We must determine what effect, if any, comparative fault has had upon rescue cases in the context of the two aforementioned situations and, if the jury in this case was properly instructed thereunder.

Gustafson v. Benda, 661 S.W.2d 11 (Mo. banc 1983), supplanted the draconian rule of contributory negligence with a more equitable system of comparative fault. Id. at 16. The purpose was to eliminate the inherent inequity in a legal doctrine that irrationally imposed total responsibility upon one party for the consequences and conduct of both parties. Earll v. Consolidated Aluminum Corp., 714 S.W.2d 932, 936 (Mo.App. 1986). Accordingly, where there is evidence that the conduct of both parties combined and contributed to cause damage, the fact finder should not be precluded from comparing the respective contributions toward such causation made by each party. Id.

Missouri, has judicially adopted the "pure" form of comparative fault. Lippard v. Houdaille Industries, Inc., 715 S.W.2d 491, 493 (Mo. banc 1986); H. Woods, Comparative Fault, § 1:11 at 26-28 (2nd ed. 1987). Under the "pure" form of comparative fault, a plaintiff's negligence that concurs with that of the defendant does not relieve the defendant from liability. See Am.Jur.2d, Comparative Negligence, § 2, at 4-5 (New Topic Service 1977). It merely diminishes the amount of damages that the plaintiff can recover. Id. "Pure" comparative fault seeks to compensate an injured plaintiff so long as his portion of the combined negligence of the parties is less than 100%. Id. § 10, at 13-14.

We find no logical reason why, under our "pure" comparative fault system, a rescuer's negligence in contributing to create a situation of peril that summons him to the aid of the one imperiled should not be compared to the negligence of the defendant in contributing to the creation of the peril. The law places a premium on the preservation of human life and one who attempts the laudable act of saving life, even though he is partly responsible for having endangered it, should not be denied recovery.

Likewise, we see no justification for not comparing the negligence of a rescuer who has crossed the line of rash or reckless, in his attempt at rescue or in the manner in which he carries it out, so long as his portion of the combined negligence of the parties is less than 100%. The "pure" comparative fault rule is applicable regardless of whether the plaintiff's negligence is characterized as ordinary or aggravated. See Am.Jur.2d, Comparative Negligence, § 11, at 14 (New Topic Service 1977). Indeed, this is the rule that obtains in the majority of jurisdictions that have considered the question. See, e.g., Blake v. Moore, 162 Cal.App.3d 700, 208 Cal.Rptr. 703, 708 (1984); Estate of Starling v. Fisherman's Pier, Inc., 401 So.2d 1136, 1136-38 (Fla.App. 1981); Burlington Northern Railroad Co. v. JMC Transport, Inc., 567 F. Supp. 389, 393 (N.D.Ill. 1983) (applying Illinois law); Salster v. Singer Sewing Machine Co., 361 F. Supp. 1056, 1060 (N.D. Miss. 1973) (applying Mississippi law). See also Yazoo M.V.R. Co. v. Williams, 114 Miss. 236, 74 So. 835.

This result is consistent with the Gustafson decision. In Gustafson, the Uniform Comparative Fault Act (UCFA) was appended to and made a part of the opinion. The direction by the court was to apply the Act to appropriate cases "insofar as possible." Gustafson, 661 S.W.2d at 15. The Act, however, is merely to serve as a guide. It was not the purpose of Gustafson to enact the Uniform Act as a virtual statute of the State of Missouri, establishing substantive principles for every eventuality. Lippard v. Houdaille Industries, Inc., 715 S.W.2d 491, 492-93 (Mo. banc 1986).

Using the UCFA as our guide, we conclude that the concept of "fault" includes "acts or omissions that are in any measure negligent or reckless toward the person or property of the actor or others." Unif. Comp. Fault Act § 1(b), 12 U.L.A. 39 (Cum.Supp. 1987) (emphasis ours). See also Comment, Comparative Fault in Missouri, 50 Mo.L.Rev. 141, 150 (1985). "`In any measure' is intended to cover all degrees and kinds of negligent conduct." Unif. Comp. Fault Act § 1 comment, 12 U.L.A. 40 (Cum.Supp. 1987). Thus, under the UCFA, the rescuer whose conduct is rash or reckless, either in attempting the rescue or in the manner of rescue falls within the coverage of the Act. Such conduct should be considered in proportionately diminishing the amount awarded, but should not bar recovery. See Unif. Comp. Fault Act § 1(a), 12 U.L.A. 39 (Cum.Supp. 1987).

A review of the rescue cases from other jurisdictions that have adopted comparative fault supports this conclusion. In Cords v. Anderson, 80 Wis.2d 525, 259 N.W.2d 672 (1977), two plaintiffs were injured while attempting to rescue a companion who had fallen into a canyon. The Wisconsin Supreme Court, in applying comparative negligence to the rescue doctrine held:

that a rescuer is not negligent where the rescue, although dangerous, is not unreasonable or unreasonably carried out. In a comparative negligence jurisdiction such as Wisconsin, if the trier of fact finds that the rescue is unreasonable or unreasonably carried out the factfinder should then make a comparison of negligence between the rescuer and the one whose negligence created the situation to which the rescue was a response.

Id. 259 N.W.2d at 683.

In Ryder Truck Rental, Inc. v. Korte, 357 So.2d 228 (Fla.App. 1978), a police officer was injured while aiding the victim of an automobile accident. the court analyzed the effect that the adoption of comparative negligence had upon the rescue doctrine as follows:

Now that Florida has abolished contributory negligence [as an absolute defense], the rescue doctrine is no longer needed to allow a rescuer to recover in spite of his contributory negligence, but there is no logical reason why the principles of comparative negligence should not apply in a rescue case. We therefore hold that when the plaintiff in performing a rescue is himself negligent, he should recover only that portion of the entire damages sustained by him as the defendant's negligence bears to the combined negligence of both the plaintiff and the defendant.

Id. at 230.

In Sweetman v. State Highway Department, 137 Mich. App. 14, 357 N.W.2d 783 (1984), the plaintiff attempted to aid a motorist that had skidded on ice and collided with a bridge guardrail. The plaintiff alleged that the defendant was negligent through the improper design and construction of the highway and through ineffective maintenance. The plaintiff was injured while trying to warn approaching traffic of the accident on the bridge. The trial court found the plaintiff 75% negligent and awarded her $250,000, representing $1,000,000 in damages adjusted for her own negligence. Both plaintiff and the defendant appealed. In addressing the issue of comparative negligence as it bears on the rescue doctrine the court stated: "we perceive no harsh result from the application of comparative negligence principles to rescue cases." Id. 357 N.W.2d at 789. The court went on to set forth the duty of the trier of fact in rescue cases under comparative negligence.

Where a plaintiff suffers an injury during the scope of a rescue, the trier of fact must first inquire whether a reasonably prudent person would have acted as the plaintiff did under the same or similar circumstances. In making this determination, the trier of fact must, as in all negligence cases, balance the utility of the actor's (in this instance rescuer's) conduct against the magnitude of the risk involved.

. . . . .

If it is found that the rescuer did not act reasonably in carrying out his mission, i.e., he was negligent, he should recover only the portion of the entire damages sustained by him as the defendant's negligence bears to the combined negligence of the plaintiff and the defendant.

Id. (citations omitted).

In Pachesky v. Getz, 352 Pa. Super. 505, 510 A.2d 776 (1986), the plaintiff was injured during an attempt to assist an intoxicated motorist who appeared to have been overcome by carbon monoxide fumes. The jury found that 20% of the plaintiff's injuries were attributable to her causal negligence. On appeal, the plaintiff alleged that the trial court erred in refusing to instruct the jury that when a plaintiff is injured while attempting to rescue, the plaintiff cannot be found comparatively negligent unless the plaintiff's actions were rash or wanton. The trial court instructed the jury in pertinent part:

Now, members of the jury, there is a theory of law which you must consider in your deliberations, and that is where a person places another person or himself, where he negligently places another person or himself in a position which in the natural and normal course of events invites rescue, he is liable to the person who comes to his rescue for any injury which that person may suffer in the attempted rescue, provided, of course, that the person who attempts the rescue acts reasonably in attempting to effect that rescue.

Id. 510 A.2d at 780. The Supreme Court of Pennsylvania, in reviewing the case law as to the effect of comparative negligence on the rescue doctrine, concluded:

When a plaintiff performs a rescue in a reasonable manner, he/she is entitled to full recovery from the negligent defendant for all damages occasioned thereby. However, when a plaintiff acts unreasonably, that is, in a negligent fashion, in performing a rescue, the relative causal negligence of the parties should be apportioned....

Id. at 783. In addressing the alleged instructional error, the court found "that the court below was eminently correct in directing the jury to allow for [the plaintiff's] full recovery if the jury found her, as the rescuer, to have acted `reasonably.' The jury, as fact finder, found otherwise and, in accordance with the further instructions of the court below, the jury appropriately assigned to [the Plaintiff] a percentage of the causative negligence." Id.

In Zimny v. Cooper-Jarrett, Inc., 8 Conn. App. 407, 513 A.2d 1235 (1986), the plaintiff's decedent was struck by a vehicle while attempting to help her mother out of their car after being involved in a multi-vehicle accident. The Supreme Court of Connecticut examined the rescue doctrine in light of comparative negligence and held "that it is not negligence for a plaintiff to make a reasonable attempt to save an individual from peril.... Where, however, the attempt is unreasonable, we find that a result different from that obtained under a contributory negligence standard should apply. The rescuer should not be absolutely barred from recovering but, rather, the amount of his recovery should be reduced by the degree to which he is comparatively negligent." Id. at 1243.

We are persuaded by the forceful reasoning in the foregoing cases. Under the rule of comparative fault in Missouri, a person who sees another in imminent peril created by the negligence of defendant will not be charged with negligence in risking his or her own life or serious injury in an attempt to rescue, provided he or she does not act recklessly or rashly. In other words, it is not negligence for a plaintiff to expose himself to danger in an attempt to save a third person from harm created by another, unless the effort to save is a rash or reckless one, or the plaintiff acts rashly or recklessly in the course of it. If the defendant establishes that the rescuer's actions were reckless or rash, the rescuer is negligent.

If it is found that the rescuer was negligent, the trier of fact should compare his or her negligence with that of the one whose negligence created the situation to which the rescue was a response. In awarding damages, the rescuer is only entitled to recover the portion of the entire amount of damages sustained that are attributable to the defendant's negligence in creating the perilous situation.

Addressing plaintiffs' claim of the erroneous omission of their tendered instructions on the rescue doctrine, our standard of review is delineated by the well settled proposition that "a party is entitled to have his own theory of the case submitted to the jury by instructions, provided the instructions are within the pleadings and evidence and are correct in both form and substance." Orloff v. Fondaw, 315 S.W.2d 430, 433 (Mo.App. 1958). The failure to submit an instruction to which a party is entitled is error, and will warrant reversal when the merits of the action have been materially affected. Id. at 434. Of course, in reviewing the evidence to determine if it will support an instruction, we view it in the light most favorable to the proponent of the instruction.

To maintain an action premised on the rescue doctrine, the plaintiff must allege that the negligence of the defendant endangered the safety of another, and that the plaintiff sustained injuries in an attempt to save the other from injury. National Dairy Products Corp. v. Freschi, 393 S.W.2d 48, 57 (Mo.App. 1965). Plaintiffs' second amended petition alleged the following: that defendant Sverdrup, SPA, STI, and Dr. Flanigan "had authority over and were responsible for the design, testing, and supervision of the GROW project;" that Sverdrup, SPA and STI had "unreasonably and negligently designed [and tested] the GROW project generally, and the feed system thereof specifically;" and that the design permitted toxic gases to accumulate in the feed bin and individual access into said bin was reasonably foreseeable. Plaintiffs alleged generally that on October 1, 1980, Kurt Holekamp, a student of UMR, entered the feed bin at the GROW project to dislodge sawdust that had adhered to the interior walls thereof; that the environment inside the bin was then composed of toxic gases; that upon entering the bin, Kurt Holekamp was overcome by carbon monoxide gases and lapsed into unconsciousness; "[t]hat Donny Lee Allison entered the [feed bin] to aid Kurt Holekamp, whereupon he also lapsed into unconsciousness." These allegations are sufficient to support an action based on the rescue doctrine. We have heretofore stated the evidence fully, and without now again quoting its different aspects, suffice to say, it was sufficient to submit an instruction on the rescue doctrine to the jury.

It is clear that the merits of the instant action were materially affected by the refusal of the trial court to clearly instruct on the rescue doctrine. Instruction No. 6, which was given over plaintiffs' objection, completely omits reference to STI's alleged negligence in creating the peril to Kurt Holekamp to which Allison responded. Plaintiffs' tendered instruction contained such a reference and, indeed, so did STI's comparative fault Instruction No. 9.

In failing to include a charge requiring a finding that Kurt Holekamp had been imperiled by the negligence of STI, Instruction No. 6 ignored the basic theory of plaintiffs' case and the rescue doctrine, i.e., that the proximate cause of the injury to the rescuer is the defendant's negligence in creating the peril which induced the rescuer to act. Lowrey, 689 S.W.2d at 627. It is obvious that, under plaintiffs' theory of the case, Allison would not have died had Kurt Holekamp not been imperiled because of STI's alleged negligence. Instruction No. 6 misled and confused the jury. It did not inform them that Allison was a rescuer. It charged the jury that it could only find for plaintiffs if Allison's death was a "direct result" of STI's negligence and failed to inform the jury that plaintiffs could recover if STI's negligence had placed Kurt Holekamp in a position of peril to which Allison had responded. We conclude that failure to instruct on the rescue doctrine was prejudicial and the cause must be reversed as to plaintiffs' claims against STI.

We conclude likewise, that the giving of Instruction No. 9 was erroneous and clearly affected the merits of the case. We are of the opinion that Instruction No. 9, the comparative fault instruction, confused the jury with respect to the standard of care required of Allison.

STI by its answer asserted that "whatever injuries and damages plaintiffs sustained were caused by or contributed by the negligence of plaintiffs' decedent, Donny Lee Allison." STI's proof at trial tended to show that Mr. Allison was negligent in placing Kurt Holekamp in a position of peril and in the manner he attempted to rescue Holekamp. Thus, Instruction No. 9 instructed the jury to assess a percentage of fault to plaintiffs if Mr. Allison was negligent in: "directing Kurt Holekamp to enter the feed bin contrary to instructions;" "failing to properly supervise Kurt Holekamp when Kurt Holekamp was working on the feed bin;" and "failing to purge the feed bin with air before Kurt Holekamp entered the feed bin." Clearly, Allison's alleged negligence in the three foregoing respects goes to the negligent creation of peril. However, Instruction No. 9 also instructed the jury to assess a percentage of fault if "Donny Allison failed to follow feed bin entry instructions when he entered the feed bin." Allison's alleged negligence here plainly refers to a rescue attempt.

As we have discussed, two different standards of conduct govern a rescuer's actions with regard to his creation of peril and his rescue attempt. We are of the opinion that including both in the same instruction is erroneous, especially where the jury is not advised that a rescuer is not negligent in exposing himself to danger in an attempt to save another from harm, unless the effort to save is a rash or reckless one, or the rescuer acts rash or reckless in the course of it. Instruction No. 9 failed to inform the jury of the differences in care required of a rescuer in creating peril and in attempting rescue. Moreover, it did not instruct the jury that no comparative fault may be assessed to Allison for his rescue attempt unless his conduct was rash or reckless.

In plaintiffs' second point, they maintain that the trial court erred in directing verdicts in favor of Sverdrup and SPA. In reviewing the alleged error we "must consider the evidence and all reasonable inferences therefrom in the light most favorable to plaintiff[s] in order to determine whether plaintiff[s] made a submissible case against [either] of the defendants." Rustici v. Weidemeyer, 673 S.W.2d 762, 765 (Mo. banc 1984). We reject all unfavorable inferences, and disregard defendants' contrary evidence. See Beshore v. Gretzinger, 641 S.W.2d 858, 862 (Mo.App. 1982).

Plaintiffs maintain that the "evidence affirmatively illustrated the involvement of [Sverdrup] and [SPA] with the design of the feed handling system of the GROW project and was, therefore, clearly sufficient to require submission of plaintiffs' claims against these defendants." We note that substantial evidence is required to establish a submissible case. Fujita v. Jefferies, 714 S.W.2d 202, 206 (Mo.App. 1986). A mere scintilla of evidence is insufficient and forced or violent inferences should not be indulged. Jones v. Garney Plumbing Co., 409 S.W.2d 637, 645 (Mo. 1966).

There is no direct evidence that Sverdrup or SPA had any part in the actual design of the feed system in which Allison died. However, plaintiffs maintain that they proved that Sverdrup and SPA were involved in the design through legitimate "inference ... that these defendants were involved in the design of the GROW project facility in which Kurt Holekamp and Donny Allison met their deaths."

Plaintiffs maintain that they proved Sverdrup was involved in the design of the feed system by way of statements contained in two documents. The first is the report submitted to the Department of Energy in 1978. The report stated that "Mr. [Ron] Williams, [of] Sverdrup Corporation, will serve as senior investigator to provide technical interface between UMR and the Sverdrup Corporation, to provide engineering services in the implementation of the program as a sub-contractor to the University of Missouri at Rolla." The second document was prepared by the Sverdrup Corporation and it related the following:

Work was completed in 1980 on the development of a biomass conversion process at the University of Missouri at Rolla. There, Sverdrup engineers developed and tested a fluidized bed wood gasifier, capable of handling 24 tons of wood per day. Results of the two-engineer project will be compiled to describe the feasibility of converting wood to low and medium BTU gas on a commercial scale.

These documents are simply insufficient to support an inference that Sverdrup was responsible for the design of the feed system. An inference is a logical a priori conclusion drawn by reason from proven or admitted facts. It is more than, and cannot be predicated on, mere surmise or conjecture. It is not a possibility that a thing could have happened or an idea founded on the probability that a thing may have occurred. Smith v. Seven-Eleven, Inc., 430 S.W.2d 764, 769 (Mo.App. 1968). Here, as to the Department of Energy Proposal, plaintiffs admit that Ron Williams was employed by SPA, a subsidiary of Sverdrup, and not Sverdrup. In their brief, plaintiffs designated "Ron Williams [as] the project manager at [SPA]" and, indeed, all of the other evidence confirmed this.

As to the document published by Sverdrup and quoted above, it must be remembered that two subsidiaries were involved (STI and SPA) and that, unlike our opinion, see supra, Sverdrup was not limited in the document to refer only to Sverdrup Corporation. Thus, the reference in the document could relate to STI or SPA or any one of the other Sverdrup companies. To conclude that this evidence implies that Sverdrup designed the feed system is to resort to speculation and conjecture. It does not afford a sufficient basis for a logical a priori conclusion drawn by reason from proven facts. See Howard v. Kysor Industrial Corp., 729 S.W.2d 603, 605-06 (Mo.App. 1987).

Plaintiffs contend that they made a submissible case against SPA through three individuals: Dr. Dave Fenton, Dr. Ron Williams, and Dr. Virgil Flanigan. Plaintiffs focus on Fenton's periodic conversations with Dr. Emmons, Dr. Omurtag, Dr. Welch, and Dean Halligan during 1978. However, there is nothing in the record to indicate that these conversations had anything to do with the actual design specification for the feed system used in Phase II of the GROW project during 1980, i.e., the system in which Allison died.

Ron Williams' alleged involvement in the design is even more attenuated. Plaintiffs point out that in the Spring of 1978, Williams travelled to Colorado to inspect the Coors equipment and later accompanied Fenton to UMR to meet with Dean Halligan and Dr. Emmons. Such activity is not indicative that Williams in any way participated in engineering or designing the feed system that proved fatal to Allison two years later.

Finally, plaintiffs point to Dr. Flanigan to link SPA with the design of the feed system. Flanigan, a professor at UMR, was employed by SPA during the summer of 1978. During that time he worked on the Department of Energy proposal. However, Flanigan testified, and there was no evidence to the contrary, that his work on the proposal was done at the behest of Dean Halligan and not SPA and that his work on the proposal was not part of his duties at SPA. This was corroborated by the uncontradicted testimony of Dr. Fenton who said that Flanigan's work on the proposal was considered by SPA as something he did on his own free time. It is true that Fenton testified that SPA had a long term financial interest in the GROW project, but again, this is short of establishing any sort of inference that SPA designed the feed system.

We conclude that the trial court providently granted the directed verdicts in favor of Sverdrup and SPA and those judgments are affirmed. However, because of the instructional error in plaintiffs' case against STI, that judgment must be reversed and the cause remanded for proceedings consistent with this opinion.

Affirmed in part, and reversed and remanded in part.

GAERTNER, P.J., and STEPHAN, J., concur.