The plaintiff, as surviving wife, brought this action under the North Dakota Death by Wrongful Act statute, Chapter 32-21, North Dakota Century Code. Her husband, a licensed master plumber, at the time of his death was an employee of Pfaff Sheet Metal, an Enderlin firm which had been the successful bidder on part of the work in constructing an addition to the Jamestown High School. The deceased was acting as foreman for his employer on the project.
The accident occurred early in the morning of November 3, 1967, in a sump room in the basement of the new addition. This room was located just off the swimming pool area. The plaintiff’s husband had descended into the sump to assist another man, and both had lost consciousness. The cause of their difficulty is in dispute. An employee of another contractor went to their assistance and he, too, was overcome but later was revived. Rescue attempts to revive the others were unsuccessful.
The defendant Otter Tail operated a gas-distributing system in the city of Jamestown, and Jamestown High School was one of its customers. During the summer and early fall of 1967, while the school addition was under construction, the deceased, on several occasions, had told one of the architects on the project that he had detected an odor of some kind in the sump and that he was having difficulty breathing while working in that place. The architect thereupon requested Otter Tail to conduct tests for the presence of gas in the sump. On September 19, 1967, employees of Otter Tail located a gas leak in a service line leading from the main line in the street adjacent to the high school. They dug down to the leak, cut the service line, and capped it. This leak was located approximately fifty to sixty feet from the nearest wall of the school building. Tests made at that time resulted in no readings which disclosed any gas at any point farther than ten feet from the point of the leak. No other leaks were ever discovered in the area of the high school, either before or after September 19.
Odors persisted in the sump, however, and on October 6, 1967, almost one month before the date on which the plaintiff’s husband died, a gas expert from Otter Tail, with an assistant, tested the air in the sump to determine whether it was explosive. Using a mine-safety device, these men received a 100 per cent explosive reading. The éxpert asserts that he informed the architects of this reading, which assertion is denied by the architects. At the trial, this expert testified that he had conducted two separate tests on the 6th day of October 1967 and had received the 100 per cent explosive reading on the first test. A negative reading was received on the second test made later in the day. Neither Otter Tail nor the architects at any time notified the fire department or the State Fire Marshal’s office of the difficulties encountered in the sump.
Following the accident on November 3, further tests of the atmosphere in the sump and at other locations at the school were made by Otter Tail and by persons from the Workmen’s Compensation Bureau and the Jamestown fire department. These tests included those made with mine-safety-appliance explosi meters, which tests revealed the presence of 100 per cent explosive atmosphere in the sump. On November 4, the day after the accident, the State Toxicologist, Richard Prouty, took samples of air from the sump. He was not advised that the gas being distributed to Jamestown residents by Otter Tail had a propane base. The tests which he conducted did not disclose explosive gas, and he attributed the death of the plaintiff’s husband to a lack of oxygen. *194The toxicologist, however, did not test for the presence of propane gas until just prior to the trial of this action. The results of such subsequent tests were excluded by the trial court. The failure to admit these subsequent tests relative to the presence of propane gas in the sump is claimed as error by the appellant. Readings which disclosed a high explosive content resulted from various tests taken of the sump atmosphere following the accident. Such readings persisted until the walls of the sump were sealed and a ventilating system was installed. The trial court sustained Otter Tail’s objection to evidence showing these changes, and another of the issues presented on this appeal is whether the court properly excluded evidence showing the alterations made subsequent to the accident.
The record further discloses that for some months prior to the date of the accident, Pfaff Sheet Metal, the employer of plaintiff’s husband, had maintained on the premises a 100-pound tank of propane gas, several 20-pound tanks of propane gas, and several 1-pound hand tanks of this substance. The propane in these tanks was from a source other than Otter Tail. The gas in these tanks was used by the deceased and other employees of Pfaff in connection with their work on the project. On several occasions prior to the accident, the deceased’s employer had talked with the deceased about the odors of which he was complaining, and the employer identified such odors as being sewer gas.
The trial court ordered the action dismissed against Otter Tail at the conclusion of the plaintiff’s case, but permitted the case against the architects, the defendants Horton, to go to the jury. The jury returned its verdict for the defendants, and judgment was entered thereon. This appeal is by the plaintiff from the judgment dismissing the complaint against Otter Tail and from the judgment entered on the jury’s verdict dismissing the case against the defendants Horton.
Numerous assignments of error, including rulings of the trial court on the admission or the exclusion of evidence, the failure of the trial court to give certain requested instructions, the entering of pretrial orders, and the directing of a verdict in favor of the defendant Otter Tail put in issue many of the trial court’s solutions of legal issues arising before and during the trial. All of these specifications will be considered, although not in the order in which they are presented by the appellant in her brief.
The first specification of error which we will consider is the specification based upon the trial court’s refusal to allow the appellant to read into the evidence the depositions of one Solee and one Poole, both employees of the defendant Otter Tail. The appellant asserts that these two witnesses were superintendents or managing agents of the defendant Otter Tail, a corporation, and that the appellant therefore should have been permitted to read their depositions into evidence as substantive proof in support of the plaintiff’s case, even though they both were available as witnesses at the trial.
The use of pretrial depositions of witnesses and parties at the trial of an action is provided for in Rule 32 of the North Dakota Rules of Civil Procedure. Subdivision (a) (3) of that rule permits any party to use the deposition of a witness, whether or not such witness is a party, for any purpose, provided it first is shown that such witness is not available as a witness at the trial, for reasons set forth. Subdivision (a) (2) permits an adverse party to use the deposition of a party or of anyone who at the time of giving the deposition was the superintendent or managing agent of a private corporation which is a party to the action, “for any purpose.”
The limitation on the use by an adverse party of the deposition of a party or of a managing agent of a corporation which is a party is a matter of first impression in this court. Our Rule 32, however, is the *195same as Rule 32 of the Federal Rules of Civil Procedure, and we find that the Federal courts frequently have interpreted this rule. Federal decisions have given Rule 32(a)(2) a liberal construction and have held that a deposition of a party or of a managing agent of a corporation which is a party may be introduced by the adverse party as part of his substantive proof, without regard to whether such party or managing agent is available at the time of trial. Pfotzer v. Aqua Systems, Inc., 162 F.2d 779 (2d Cir. 1947); Pursche v. Atlas Skyscraper & Engineering Co., 300 F.2d 467 (9th Cir. 1962); Community Counseling Service, Inc. v. Reilly, 317 F.2d 239 (4th Cir. 1963); Zimmerman v. Safeway Stores, Inc., 133 U.S.App.D.C. 342, 410 F.2d 1041 (1969); Pingatore v. Montgomery Ward & Co., 419 F.2d 1138 (6th Cir. 1969).
We believe that the construction placed on this portion of Rule 32 by these decisions is the correct interpretation of the rule, and we adopt their reasoning in interpreting our rule, which is the same.
Having concluded that Rule 32(a)(2), North Dakota Rules of Civil Procedure, allows an adverse party to use the deposition of a party or of a managing agent of a corporation which is a party as part of his substantive proof, without regard to the availability of the party or the managing agent as a witness at the trial, we find that it was error for the trial court to deny the plaintiff the right to read the depositions into evidence, assuming, of course, that the witnesses were managing agents of the defendant Otter Tail, as asserted by the appellant. See 4A Moore’s Federal Practice, 2d Ed., Par. 32.04.
While we hold that it was error for the trial court to refuse to allow the appellant to read these depositions into evidence, we also find that in this case such error was harmless. This conclusion is based upon the comparison of the testimony of these witnesses as given at the trial and their testimony as given in their depositions. This comparison discloses that the attorney for the plaintiff, during the course of his extensive cross-examination, was able to read into evidence all of the pertinent parts of their depositions. We reach the conclusion that the practical effect of the plaintiff’s use of the depositions, on cross-examination, was the same as it would have been had plaintiff been permitted to read the depositions into the record as substantive proof. The trial court at no stage of the proceedings limited consideration of these depositions by the jury merely for impeachment purposes.
Although we hold that the deposition of a party, or a managing agent of a corporation which is a party, may be used as substantive proof of the adverse party’s case without regard to whether such persons are available as witnesses at the trial, we believe that use of the depositions may, as a practical matter, be limited by the trial court in the exercise of its sound discretion. We believe that the trial court may require counsel to specify what parts of such depositions it deems to be relevant and limit the reading thereof to those parts, subject to objection by opposing counsel, pursuant to Rule 32(a)(4), North Dakota Rules of Civil Procedure. See Pfotzer v. Aqua Systems, Inc., supra.
We next will consider the specification in which the appellant asserts that the trial court erred in refusing to allow into evidence the results of certain tests conducted by the State Toxicologist, Richard Prouty. This witness testified that he arrived at the scene of the accident in Jamestown the morning following the accident. He took samples of air from near bottom of the sump, in which the plaintiff’s husband had died the previous day. At the time he recovered these samples, Prouty was not aware of the fact that gas being distributed by Otter Tail to its Jamestown customers was a propane and air mixture, but assumed it to be a methane and air mixture.
*196While unaware of the fact that a propane mixture, and not a methane mixture, was the gas being distributed to Otter Tail’s customers in Jamestown, Prouty, on November 6, analyzed the air samples taken from the sump. This analysis was made by passing the samples through a gas chromatograph. This is a device which separates the various components of a gaseous mixture by chemical and by physical means. The device is filled with various chemicals and the gas mixture then is forced to pass through the machine in contact with these chemicals. Because of different physical and chemical properties of each ingredient of the gas mixture, each component is retained in the device for a different length of time. The difference in the time it is retained is recorded on graph paper.
Before the gas mixture which is to be analyzed is passed through the chromato-graph, quantities of various known gases for which the gas mixture is to be tested are passed through the gas chromatograph and the various retention times for these gases are recorded. These retention times then become the standard with which the retention time for the various components of the gas mixture are compared. When the retention time for one of the components of the gas mixture is the same as that of one of the standard gases, the analyst may conclude that the component gas is, in fact, the same as the standard gas.
This procedure was followed by Dr. Prouty. Before forcing the gas mixture from the sump through the gas chromato-graph, he processed several standard gases, with the exception of propane, through the instrument. As a result of Dr. Prouty’s comparison of the retention times for the components of the gas mixture from the sump to the retention times for certain standard gases — not, however, including propane — he concluded that the sample of the gas mixture from the Jamestown sump contained eight per cent carbon dioxide, five per cent oxygen, and eighty-seven per cent nitrogen. This conclusion and the graph papers which supported it were received in evidence at the trial.
A short time before the trial of the action, Dr. Prouty learned that a propane and air mixture was the commercial gas being distributed by Otter Tail in Jamestown. On acquiring this information, he forced through the gas chromatograph a sample of propane taken from his laboratory in Fargo. He also ran a sample of nitrogen through the gas chromatograph and the result of the test thus conducted was that the propane gas had the same retention time as did nitrogen. This information was offered in evidence by the plaintiff and was excluded by the trial court. Thereafter, an offer of proof was made, and in the offer of proof Dr. Prouty testified that the test indicating the identity of retention time for propane and for nitrogen was made under circumstances which were the same as those existing at the time that his original test of the gas mixture from the sump was made. Prouty further testified that in the light of this identity of retention time for propane and nitrogen, he would be unable to support his conclusion, based on the original test, that air in the sump contained eighty-seven per cent nitrogen. Instead, he stated that, based on the results of the original test and the one made a short time before the trial, he would be unable to conclude whether propane or nitrogen, or a mixture of the two gases, was present in the air samples taken from the sump.
Experiments, demonstrations, and tests of various kinds frequently serve to give the jury knowledge which is important to its determination of the issues and which it could not otherwise readily obtain. This court has held that evidence of relevant tests is admissible where the tests are shown to have been made under circumstances which assure the reliability and probative value of such tests, but that admission of evidence of such experiments and tests is primarily within the sound discretion of the trial court. The exercise of this discretion will not be disturbed on ap*197peal unless an abuse of discretion is shown. Larson v. Meyer, 135 N.W.2d 145 (N.D.1965); Fisher v. Suko, 111 N.W.2d 360 (N.D.1961); Larson v. Meyer, 161 N.W.2d 165 (N.D.1968); Armstrong v. Miller, 189 N.W.2d 688 (N.D.1971). This court has held that the foundation testimony offered in support of the admission of the results of tests must establish the following conditions before the results of such tests are admissible in evidence: (1) The person making the test must be competent; (2) The conditions surrounding the test must be substantially the same as those prevailing at the time of the occurrence to which they relate; (3) The results must be shown to be relevant to an issue in the case; and (4) The test must have been made honestly and fairly. There is no precise way of determining whether or not the foundation testimony establishes these four conditions. This determination must depend upon the particular factual situation presented by each case, and therefore must rest upon the exercise of sound discretion of the trial court. Larson v. Meyer, 161 N.W.2d 165 (N.D.1968).
However, in exercising its discretion, the trial court should always be aware of the principal object to be achieved by admission into evidence of results of tests, and that is the ascertainment of truth with reference to the existence or non-existence of facts in controversy. 32 C.J.S. Evidence § 587, pp. 715-717.
The foundation testimony offered by the plaintiff in support of the admission into evidence of the results of the tests conducted on November 6 by Dr. Prouty indicated that Prouty was competent to conduct this type of test and also to perform the type of test which he carried out shortly before the trial. We therefore find that the first condition for admission of the results of a test was established by the foundation testimony offered by the plaintiff. This testimony further indicates that the only circumstance of the test conducted fore the trial which was different from the circumstances surrounding the making of the test on November 6, 1967, was that the propane tested shortly before trial was not taken from the Otter Tail Power Company’s distribution system in Jamestown. This, asserts Otter Tail, totally undermines the probative value of the second test. We do not agree, however. Prouty’s tests indicate that the source of propane and the strength of the propane which is introduced into the chromatograph has absolutely no bearing on the retention time indicated on the graph paper. Therefore, the foundation testimony of Dr. Prouty indicates that the circumstances surrounding the test conducted a short time before the trial were, for all practical purposes, identical in every detail with those surrounding the earlier test. This identity of circumstances is sufficient to meet the second requirement of admission of the results of such tests. Minor variations in the circumstances go to the weight of the evidence to be given to the results, and not to the admissibility.
The testimony offered by the plaintiff further indicates that at approximately the same time that the samples were taken from the sump, the air in the sump was 100 per cent explosive, as indicated by ex-plosimeter tests. Thus, when the results of the test conducted just before the trial are viewed in the light of the fact that the air in the sump which was analyzed by Prouty was 100 per cent explosive, we are constrained to adopt the inference that propane in an undetermined amount was present in the air sample which was tested by Prouty shortly after the accident. Otter Tail asserts, however, that the results of the test just before the trial are irrelevant because they are inconclusive, and that an inference of the presence of propane is not supported by the results of the second test since the results of that test do not eliminate the possibility that some other explosive gas may have a retention time identical to that of nitrogen, and that such gas might have been present in the sump when the samples were taken.
*198We find this argument unpersuasive. The results of the test made by Prouty following the accident, which were admitted into evidence, indicate that methane is a prime component of sewer gas and has a retention time in the gas chromato-graph which is distinguishable from that of nitrogen. Otter Tail urges that the results of the second test do not support an inference that propane gas was present in the sump because the results of that test do not eliminate the possibility that paint-thinner vapors may have a retention time identical to that of nitrogen. We believe this, too, is unpersuasive. The plaintiff in this case is under no duty to eliminate every other imaginable agency which might possibly have been present in the sump and caused or contributed to the death of the plaintiff’s husband. Rather, defendants might have shown, if they had so desired and were able to do so, that some other agency was the cause of the death. Therefore, since the plaintiff in her proof need not eliminate all possibilities of other gas being present in the sump, we believe that it was error to refuse the introduction of results of the test made just prior to the trial, which was probative of the presence of propane, simply because that test, while eliminating the possibility of some other gases, does not eliminate the possibility of one.
The foundation testimony offered by the plaintiff in support of admission into evidence of the results of the test conducted by Prouty just before the trial also establishes that that test was conducted in an objective, fair, and honest manner. Consequently, all the requirements for the admission into evidence of the results of such second test had been established by the plaintiff.
Otter Tail further argues that even if the results of the second test should have been admitted according to the rules of evidence, the results of the second test and Prouty’s conclusions based thereon should not have been admitted because to do so would be contrary to the rule prohibiting a party from impeaching his own witness. This impeachment, Otter Tail asserts, would occur because the results of the second test and the conclusions drawn therefrom contradict the results of the test taken immediately after the accident, which were admitted into evidence. This contention, we believe, is not well founded. In Jacobson v. Mutual Benefit Health & Accident Assn., 70 N.D. 566, 296 N.W. 545 (1941), this court quoted with approval 6 Jones, Commentaries on Evidence, 2d Edition, Section 2432, page 4814, as follows:
“The general rule that one may not impeach his own witness must not be understood as implying that the party calling a witness is bound to his version of material facts as correct. On the contrary it is very clear that the one producing a witness may prove material facts by any other competent evidence, even though the effect of such testimony is directly to contradict his own witness; ‘and this, not only when it appears that the witness was innocently mistaken, but even where the evidence may collaterally have the effect of showing that he was generally unworthy of belief.’ The object of the inquiry is not to discredit the witness, but to prove the facts relevant to the controversy; and this should be permitted whatever the incidental result may be upon the credit of any witness. . . . ”
In the light of this statement of the law, limiting the broad sweep of the ban of the general rule that a party may not impeach his own witness, and in the light of the fact that the contradictory evidence arose because of Prouty’s mistaken belief that the gas being supplied by Otter Tail in Jamestown was a methane and air mixture, and r.ot a propane and air mixture, we conclude that the admission of the results of the second test and Prouty’s conclusions based thereon should have been admitted and that the results of the second test are not in contravention of the rule *199against impeachment of one’s own witness. The introduction into evidence of the conclusions based on the second test would simply indicate that Prouty contradicted himself because of lack of information at the time of making the original test. Refusal to admit into evidence the results of the second test and the toxicologist’s conclusions based thereon therefore was error.
We next will consider the plaintiff’s specification of error based on the trial court’s issuance of its pretrial order admonishing the plaintiff and her counsel not to mention, refer to, interrogate concerning, or attempt to convey to the jury in any manner, either directly or indirectly, the fact that the sump in which the plaintiff’s husband died was altered subsequent to his death. No mention of this fact was to be made before the jury without first obtaining permission of the court outside the presence and hearing of the jury. Another specification of error asserts that the trial court’s refusal to allow the appellant to read into evidence certain interrogatories and answers thereto concerning alterations in the design and construction of the sump subsequent to the death of the plaintiff’s husband was error. These two specifications will be considered together.
Rule 16 of the North Dakota Rules of Civil Procedure provides that in any action after issue is joined, the court in its discretion may, and upon written request of a party shall, direct the attorneys for the parties to appear before it for a conference in advance of trial to consider various matters. Among such matters are “(6) Such other matters as may aid in the disposition of the action. . . . ” The rule goes on to provide that the court shall make an order which recites the action taken at such conference, and such order, when it is entered, controls the subsequent course of the action unless modified at the trial to prevent manifest injustice.
Rule 16 of the Federal Rules of Civil Procedure, which is similar to our Rule 16, has been interpreted by the Federal courts as enabling them to issue pretrial orders resolving issues affecting the introduction and admission of evidence. United States v. Certain Tracts of Land, 57 F.Supp. 739 (S.D.Cal.1944); 3 Moore’s Federal Practice, Vol. 3, Par. 16.16, pp. 1124-1127.
We adopt the interpretation approved by the Federal courts and apply it to Rule 16 of the North Dakota Rules of Civil Procedure. Therefore, in the light of the power of the trial court to issue a pretrial order resolving the issue of introduction and admission of evidence, and in the light of the fact that the plaintiff was given timely notice of the hearing on the motion for pretrial order, which hearing was held in conjunction with the pretrial conference in this case, we conclude that the trial court properly exercised its powers under Rule 16 when it issued its pretrial order affecting evidence touching upon the fact that the sump had been altered following the death of the plaintiff’s husband. We therefore conclude that the plaintiff’s specification asserting that it was error for the court to issue such pretrial order is without merit.
Was it error for the court to refuse to allow the plaintiff to read into evidence interrogatories propounded to and answers submitted by the defendants Horton concerning such alterations subsequent to the death of plaintiff’s husband? The plaintiff, in her offer of proof, read the interrogatories and answers to the court and stated that they were not being offered as evidence purely of negligence, but to show that after the repairs were made, no further explosive readings were secured when tests were made of the air in the sump.
Evidence of repairs or alterations made by a defendant after an accident is not admissible either to prove antecedent negligence or as an admission of negligence. Huus v. Ringo, 76 N.D. 763, 39 N.W.2d 505 (1949).
This general rule of admissibility of evidence of repairs and alterations after *200an accident is subject to certain exceptions. 170 A.L.R. 8; 64 A.L.R.2d 1299. One of these exceptions is that the evidence of subsequent repairs is admissible to show the cause of the injury to the plaintiff’s husband. Thus evidence of subsequent repairs, while not admissible as evidence of defendants’ negligence, might be admissible to show the abatement of the cause of the injury after the repair or alteration. This was the purpose for the plaintiff’s offering of such evidence of subsequent repair. In her offer of proof, counsel stated that the evidence was not “offered as evidence purely of negligence but just to show that this is when Mr. Robertson stopped getting his readings.” David Robertson, assistant fire chief, helped remove the victims from the sump. He returned that afternoon to make explosimeter tests of the air in the sump, and determined that the air was 100 per cent explosive. Thus plaintiff’s offer of proof was to show when Mr. Robertson stopped getting the explosive readings. Robertson’s testimony is devoid of any statement as to when he stopped getting highly explosive readings on his explosime-ter from air in the sump. Furthermore, such interrogatories and the answers thereto failed to indicate when the subsequent repairs or alterations to the sump were made.
Therefore, this evidence has no probative value as to the establishment of the time when the explosive gas stopped entering the sump. We conclude that the trial court was justified in excluding the evidence of subsequent repairs when offered for the purpose of showing the cause of the death of plaintiff’s husband, and that such evidence was properly excluded even though it may have been admissible for certain limited purposes. The party making the offer of proof has the burden of showing the purposes for which the evidence is admissible. It is not the duty of the trial court to look beyond the purpose for which such evidence is offered to determine whether it is admissible, and the trial court may exclude evidence which, although admissible for certain purposes, is not admissible for the purpose for which it is offered. 88 C.J.S. Trial § 82-c, p. 189; S3 Am.Jur. Trial, Sec. 103, p. 92.
The next specification of error which we will consider is the denial of a requested instruction relative to the authority of the architects to issue change orders to correct conditions endangering life or property, with the prior approval of the owner, and to do so in emergency situations involving danger to life and property without the approval of the owner, and the authority to compel contractors to perform work not covered under the original plans and specifications to prevent threatened loss of life or injury to the work, and their authority to stop the work itself when such action might be necessary in the reasonable opinion of the architects to insure the proper execution of the contract. In lieu of the requested instruction, the trial court gave an instruction making such authority of the architects under the contract a question of fact for the jury by instructing the jury that if it found from the evidence that the architects had the authority indicated, and if it further found that the defendant architects, or either of them, had knowledge of conditions in the sump which were dangerous to life or property and failed to exercise the authority vested in them to correct such conditions, then the jury might consider such failure to act as negligence.
The basic differences, as we see them, between the plaintiff’s requested instruction and the instruction actually given by the district court covered by this specification are (1) that the district court’s instruction contains no instruction on the liability of the architects for any defective design of the sump in which the plaintiff’s husband died; and (2) the trial court’s instruction makes the issue of the authority of the architects under the contract to alter the conduct of the work on such project for the purpose of correcting conditions endangering life a question of fact for the jury and not a question of law. We believe the court did not err in refusing to *201give the instruction requested. An architect is not strictly liable for defective designs, as the plaintiffs requested instruction would indicate. Ressler v. Nielsen, 76 N.W.2d 157 (N.D.1956); 88 C.J.S. Trial, § 380, p. 967. Rather, an architect may be held liable for negligence only in failing to exercise the ordinary skill of his profession in designing a building. Ressler v. Nielsen, supra.
Consequently, we conclude that the appellant’s requested instruction under which the architects could have been held legally responsible for defects in the design of the sump, without also instructing the jury that the defects in such design must be found to have been the result of the architects’ negligence, was properly refused. We conclude that the trial court’s failure to instruct on the architects’ liability for negligent design is not reviewable on this appeal. A careful examination of the record indicates that counsel for the plaintiff at no time took exception to the court’s failure to instruct on negligent design, although the court made a timely request for any exceptions to the instructions. Rule 51(c), N.D.R.Civ.P.; Christensen v. Farmers State Bank of Richardton, 157 N.W.2d 352 (N.D.1968); Bartholomay v. St. Thomas Lumber Co., 148 N.W.2d 278 (N.D.1966); Julson v. Loyal Order of Moose No. 822, 140 N.W.2d 39 (N.D.1966); Chicago, M., St. P. & P. R. Co. v. Johnston’s Fuel Liners, Inc., 122 N.W.2d 140 (N.D.1963).
We would point out that in this case there was no evidence on the part of the plaintiff to support the claim of negligent design of the sump. Thus refusal to instruct the jury on negligent design would not have been error since failure to charge on an issue which is not supported-by evidence is not error. Teegarden v. Dahl, 138 N.W.2d 668 (N.D.1965); Chandler v. Hjelle, 126 N.W.2d 141 (N.D.1964); Austinson v. Kilpatrick, 105 N.W.2d 258 (N.D.1960). Before a jury would be permitted to pass on the question of negligent design, the plaintiff would have to show the standard set by the learning, skill, and care ordinarily possessed and practiced by others of the same profession. The jury would not be permitted to set up arbitrarily a standard of its own and then determine whether the conduct of the architects comes up to such standard. Paxton v. Alameda County, 119 Cal.App.2d 393, 259 P.2d 934 (1953). Since no evidence of this nature was presented in this case, there was no evidence of the standard of care to which the defendant architects were to be held, and therefore any instruction on that issue was properly omitted from the instructions given to the jury.
The plaintiff further asserts that the trial court erred when it left to the jury as an issue of fact the question of the architects’ authority under the building contract to alter the conduct of the work for the purpose of correcting conditions endangering life. Specifically, the plaintiff charges that the trial- court erred when it did not find as a matter of law that the defendant architects had the authority under the contract in this case to stop all work in the sump for the purpose of correcting any dangerous condition.
In Stetson v. Investors Oil, Inc., 140 N.W.2d 349 (N.D.1966), this court held that where the intention of the parties to a contract cannot be determined from the contract itself because the terms are somewhat ambiguous, the determination of the intention of the parties becomes a task of the trier of fact. See 17A C.J.S. Contracts § 617, p. 1250. Whether an ambiguity exists in the contract must be determined by the court as a matter of law; and when the terms of the contract are found to be ambiguous, it is within the province of the jury to determine the intention of the parties, and the jury may consider subsequent acts of the parties and other extrinsic evidence in determining such intention. Stetson v. Investors Oil, supra.
*202An examination of the contract between owner and contractor in this case indicates that there is an ambiguity as to whether the architects had the authority to order the contractor to take necessary precautions for the safety of its employees. The contract contains no provisions specifically requiring the architects to make orders to insure the safety of employees. However, it does obligate the architects to see that the contract is properly executed. In addition, the contract requires the contractor to take necessary precautions for the safety of employees. Thus the question arises as to whether or not the architects’ general duty to see that the contract is properly executed includes the specific’ duty of seeing that the contractor executes his responsibility of taking necessary precautions to insure the safety of the employees. A reference in the general conditions of the contract in this case to the architects’ duty to see that the contract is properly executed is merely a restatement of his primary responsibilities as set forth in the standard form of agreement. Since the extent of the architects’ duties is unclear, the trial court properly submitted the issue of the scope of such authority to the jury for its determination.
We now come to the plaintiff’s specification which asserts that the trial court erroneously refused to instruct the jury on willful or wanton misconduct on the part of the defendants. The trial court gave no other instruction on willful or wanton misconduct in lieu of the requested instruction, and as a result the court withdrew the issue of willful or wanton misconduct and its effect on the defendants’ affirmative defenses of contributory negligence and assumption of risk from consideration of the jury.
Having carefully examined the evidence and the record, we conclude that the trial court did not err in refusing to instruct the jury on willful or wanton misconduct. We believe reasonable men could have drawn but one conclusion from the evidence as we find it in this case on the issue of willful or wanton misconduct, and that conclusion would be that the architects were not guilty of such conduct. The sole evidence in the record which would tend to support any claim of willful or wanton misconduct on the part of the defendants Horton was the contradictory testimony of an employee of Otter Tail Power Company to the effect that he took explosimeter tests in the sump on October 6, 1967, and reported his finding of 100 per cent explosive atmosphere in the sump to Kent Horton, one of the architects. Horton thereafter failed to inform the plaintiff’s husband of the result of this test even though he had ample opportunity to do so. In order to characterize an injury as having been willfully or wantonly inflicted, it is necessary to show knowledge of a situation requiring the exercise of ordinary care and diligence to avert injury to another; ability to avoid resulting harm by ordinary care and diligence in the use of the means at hand; and the omission of such care and diligence to avert threatened danger when to an ordinary person it must be apparent that the result likely would prove disastrous to another. Titus v. Lonergan, 322 Mich. 112, 33 N.W.2d 685 (1948); 57 Am.Jur.2d Negligence, Sec. 101, p. 451. The evidence in this case indicates that while Horton may have been informed of the 100 per cent reading from the explosimeter test taken October 6, it further indicates that he was informed of the negative reading received on the same afternoon and that these were the test results upon which he relied as being conclusive of the absence of any explosive gas in the sump. On the basis of this evidence, which indicated that he believed that there was no explosive gas in the sump, it cannot be concluded that the architect had knowledge of a situation which required him to warn Van Ornum in the exercise of ordinary care and diligence. The record indicates that the architect repeatedly did warn the plaintiff’s husband of the danger of working in subgrade sumps. Therefore, no reasonable man can *203conclude that the architect failed to use reasonable care and diligence to avert a threatened danger to the plaintiff’s husband.
We conclude that the instruction on willful and wanton misconduct on the part of the architects was properly refused by the court.
The final specification of error which we will consider asserts that the trial court erred in ordering, at the close of the plaintiff’s case, a directed verdict in favor of Otter Tail. In view of the evidence which was introduced by the plaintiff and the evidence which we have found heretofore to have been erroneously excluded by the trial court, we conclude that the trial court erred when it directed a verdict in favor of Otter Tail. Otter Tail asserts that its motion for a directed verdict was properly granted. This assertion is based upon what Otter Tail claims to be a total lack of evidence to support the plaintiff’s claim from which the jury could have concluded that Otter Tail’s propane gas was present in the sump at the time of the accident. Otter Tail argues that the evidence introduced by the plaintiff would equally support an inference that paint thinner or sewer gas was present in the sump, and' that the directed verdict therefore was appropriate.
We do not agree. Testimony of the State Toxicologist to the effect that he tested for methane, a prime constituent of sewer gas, and got negative results, would support an inference that sewer gas was not present. In addition, testimony of the assistant fire chief of Jamestown to the effect that in his experience sewer gas has never registered 100 per cent on an explo-simeter, as did the gas in the sump on the day of the accident, and also evidence that tests conducted by the North Dakota Workmen’s Compensation Bureau indicated an absence of hydrogen sulfide, a prime constituent of sewer gas, would support an inference that sewer gas was not present.
Otter Tail further asserts that paint thinner may have been present in lethal quantities in the sump, but we find this contention to be without merit. There is little evidence of the presence of vapors of paint thinner in the sump. This contention of Otter Tail becomes insignificant when viewed in the light of what we consider to be substantial evidence of the presence of propane gas in the sump on the day of the accident. This substantial evidence of the presence of propane gas arises, as pointed out heretofore, viewing the State Toxicologist’s second test in the light of the results of explosimeter tests indicating the presence of a 100 per cent explosive gas atmosphere in the sump at the time of the accident. Therefore, we conclude that the plaintiff’s case, when supplemented by the excluded testimony of the State Toxicologist as to results of his second test and their significance, supports a reasonable inference of the presence of propane gas in the sump at the time when the plaintiff’s husband died.
Otter Tail further asserts that even if the plaintiff’s evidence does support an inference of the presence of propane, the evidence does not support a reasonable inference that the propane present in the sump was that of Otter Tail; rather, that the evidence supports the inference that the propane came from a bottle which was present on the construction site, which did not contain Otter Tail’s propane gas. Viewing the evidence in the light most favorable to the plaintiff, as we must do in considering whether the trial court erred in granting a directed verdict, we conclude that the plaintiff’s case does support a reasonable inference of the presence of Otter Tail’s propane. While the record does indicate that several bottles of propane gas were present on the premises, there is no evidence at all to indicate whether or not any of these bottles of propane were leaking or had ever leaked, nor whether such bottles of propane were in close proximity to the sump so as to be a possible source *204of the propane which the jury may have concluded was present in the sump. Further, the evidence does show gas leaks in Otter Tail’s transmission line in close proximity to the high school building some weeks prior to the accident, and that the ground between the school and the site of the leak was covered by a hard-surface street and sidewalk. All of this, we believe, would support a reasonable inference by the jury that propane gas in the sump came from Otter Tail’s transmission lines, which were the only source of propane in the vicinity of the sump, with the exception of the bottled propane. Furthermore, the testimony and record indicate that after the accident, tests of the air in the sump were conducted by several parties and that the results of these tests varied from a zero explosive reading to a 100 per cent explosive reading, the zero reading having been received after purging the air from the sump and the 100 per cent reading having been received sometime after such purging process had stopped. We believe that this would support an inference that the propane was seeping into the sump from some surrounding area. This inference becomes more compelling when viewed in the light of the testimony indicating that at the time these tests were conducted, several persons were in the room and that the movements of such persons would disturb the regular flow of propane into the sump from any source above the ground in or near the sump.
Finally, there is testimony that the odor-ant which is present in Otter Tail’s gas may be removed from solution with the propane gas when such solution passes through soil. This fact, considered with the testimony of several witnesses to the effect that they did not detect any odor in the sump and the results of the State Toxicologist’s tests and explosimeter tests indicating that propane was present, would support an inference that the propane which the jury could have found was present in the sump actually came through the soil from Otter Tail’s transmission lines, and not from above from any bottled propane, whose odor would have been identifiable because none of the odor had been removed as a result of filtration through the soil.
We therefore conclude that the district court erred when it directed a verdict in favor of Otter Tail for dismissal of the complaint against it. However, liability for willful and wanton misconduct on the part of Otter Tail not having been established, the portion of the complaint on that issue was properly dismissed.
For reasons stated in this opinion, the judgment dismissing the complaint against Otter Tail Power Company is reversed, and the judgment based on the jury’s verdict dismissing the complaint against the defendants Horton is affirmed.
PER CURIAM.
The foregoing opinion was prepared by the Honorable Alvin C. Strutz, Chief Justice, before his death. It is adopted by the undersigned as the opinion of this court.
ERICKSTAD, C. J. and KNUDSON and PAULSON, JJ., concur.