BOSLAUGH, J.

This is an appeal in an action for damages for the wrongful death of Larry Eugene Priest. The jury returned a verdict for the defendant, Larry Lee McConnell, and the case was dismissed.

The action was brought by the personal representative of the estate of Larry Eugene Priest. He has appealed and has assigned as error (1) that the defendant's failure to timely disclose its expert witnesses violated the pretrial order; (2) that the testimony of the expert witness concerning "confabulation" was irrelevant; (3) that a proper foundation was not laid for testimony regarding the alcohol content of decedent's blood and urine samples; (4) that the trial court failed to properly instruct the jury; and (5) that the jury's use of a dictionary during deliberations was misconduct.

This action arises out of a one-vehicle accident which occurred on Highway 7, south of Ainsworth, Nebraska, on November 16, 1980. As a result of that accident, Larry Priest and Linda Lister were killed. The petition alleged that the defendant was the driver of the vehicle and that Larry Priest was a guest passenger. At trial the defendant stated that he owned the vehicle involved — a pickup truck — but that he did not remember who was driving. He testified that he had been at the Frontier Bar and the Elks Club playing pool and dancing that evening. The defendant said that he had a beer at the Frontier Bar, but did not drink at the Elks Club. He testified that he remembered leaving the Elks Club at approximately 1 a.m. and unlocking the door to his pickup. He did not recall Larry Priest or Linda Lister ever getting into his truck. The defendant stated that the next thing he remembered was lying in a ditch after the accident. He testified, however, that in the hospital, immediately after the accident, the sheriff asked him who was driving and he replied, "I guess I was." The remaining facts relate to matters concerning pretrial and trial procedures and shall be discussed as they relate to each issue.

The plaintiff's first assignment of error concerns the defendant's failure to comply with the trial court's pretrial order, which provided:

Parties are allowed to add to their witness and exhibit lists at least 30 days before trial; unless witness deposed, counsel is to provide opposing counsel with a summary statement of the testimony expected to be given by additional witnesses, and to provide copies of additional exhibits or present said exhibits for examination to opposing counsel at least 30 days before trial.

Defendant's counsel to provide the Court and opposing counsel with initial witness and exhibit lists within 60 days of this date.

Plaintiff contends that because defendant did not comply with the pretrial order and did not give timely notice of his expert witnesses, namely, Helen Waltemath and Professor Donald L. Stumpff, the trial court was in error in receiving their testimony.

The pretrial order was issued on November 16, 1982. At that time the court assessed the defendant $200 in attorney fees and $80 in mileage, payable to the plaintiff's attorneys, for not answering interrogatories in a timely manner. Answers to the interrogatories were subsequently filed.

On February 16, 1983, the defendant filed amended answers to interrogatories, identifying Dr. Joseph H. Selliken as an expert but making no mention of Waltemath or Professor Stumpff. On March 15, 1983, trial was set for May 9. On April 27, 1983, the defendant served a notice for the taking of the deposition of Helen Waltemath on April 30. Immediately thereafter, on April 29, the plaintiff filed a motion to quash the taking of this deposition, alleging that Waltemath was not listed on the defendant's witness list; that her name first appeared on a supplemental witness list received by the plaintiff on April 27, 1983; that the pretrial order indicated that additional witnesses should be added prior to 30 days before trial; and that permitting Waltemath as an additional witness would prejudice the plaintiff. The plaintiff further raised the lack of notice concerning Waltemath's deposition as one of the grounds in a motion in limine to prohibit testimony that Larry Priest was intoxicated at the time of the accident.

At a hearing on motions prior to trial, the plaintiff's and defendant's attorneys set forth different versions of a conversation between them prior to the taking of Waltemath's deposition and the plaintiff's attorneys' refusal to attend the same. The trial court sustained the motion to quash the deposition and the motion in limine insofar as the deposition was to be used to prove intoxication. The court, however, allowed Waltemath to testify in person, subject to cross-examination.

As to Professor Stumpff, his name was listed in response to plaintiff's interrogatory requesting a list of persons who have "given a written or recorded statement regarding any of the occrrences [sic] which are the subject matter of this action." He was identified as an individual who had testified at a previous criminal trial involving this accident. However, there is nothing in the transcript indicating that the defendant ever identified him as an expert witness. When asked in oral argument why the plaintiff was not given notice of Waltemath's and Stumpff's designations as expert witnesses pursuant to the pretrial order, the defendant's attorney replied that he had made a mistake. He insisted, however, that it was of no consequence, since plaintiff had not been prejudiced.

Pretrial conferences are conducted in order to simplify and narrow the issues of the case and to avoid traps and surprises. Newman Grove Creamery Co. v. Deaver, 208 Neb. 178, 302 N.W.2d 697 (1981). In the case at bar the defendant knew, well within the time limits imposed by the pretrial order, that he was going to call Waltemath and Professor Stumpff as expert witnesses at the trial. Yet Stumpff was never designated as an expert witness, and Waltemath was never identified at all until less than 2 weeks before trial. Despite defense attorney's characterization of his actions as a mistake, the result was a "trap" for the plaintiff's attorney.

This court has on previous occasions discussed counsel's failure to identify witnesses before trial. In Cardenas v. Peterson Bean Co., 180 Neb. 605, 611-12, 144 N.W.2d 154, 159 (1966), we stated:

The trial court has discretionary power to exclude the testimony of a witness whose identity is deliberately withheld in discovery under proper circumstances. The trial court would also have discretion to impose an alternative sanction to effectively protect against harm due to lack of prior knowledge of the witness, such as continuing the hearing or deferring the questioning of such a witness. The object of the rule requiring the disclosure of the names of witnesses before trial is to enable the parties to discover the truth and eliminate surprise, and, dependent on the facts, the overall policy of discovering all the truth, in some circumstances, might be more adequately served by permitting testimony after postponement until the element of surprise has been eliminated. These matters, however, are primarily within the broad discretion of the trial court.

In the case at bar the trial court quashed Waltemath's deposition and permitted an in camera voir dire of Professor Stumpff prior to his testimony at trial. While such procedures are not preferred over compliance with pretrial orders, it was within the trial court's discretion to deal with the situation in such a manner. Owing to the remedy provided, the defendant's attorney's actions do not alter the disposition of this case. That, however, does not excuse his conduct.

Of the remaining assignments of error, we first consider the relevancy of the testimony of defendant's expert witness, Professor Stumpff, regarding "confabulation." The plaintiff's pretrial oral motion in limine to exclude Professor Stumpff's testimony was sustained; nonetheless, his testimony was later received. According to Professor Stumpff, "Confabulation is to make a logical story or sense with no basis of facts." He testified that confabulation can occur when there is a blank period in a person's memory such that there is no factual basis for the recalling of an experience, but the person will try to construct a plausible explanation of his actions based on previous knowledge or past experience. In this case Professor Stumpff's testimony was presented in light of conflicting statements made by the defendant. The defendant testified that he had no recollection of who was driving on the night the accident occurred, but admitted that after the accident he was asked by the sheriff who was driving, and responded, "I guess I was."

In an in camera voir dire examination preceding his testimony at trial, Professor Stumpff stated that for an episode to be consistent with the theory of confabulation, there must have been some sort of neurological damage to the brain, either through the use of alcohol or drugs or a blow to the head. Absent a blow to the head, of which there was no evidence in this case, he stated that the plausibility of confabulation occurring depended on whether there was a level of alcohol in the body sufficient to produce "alcoholic blackout." The professor did not determine whether the defendant had consumed enough alcohol to reach that level where confabulation could occur. Furthermore, he had never examined the defendant. Professor Stumpff conceded in camera that he could not be certain that confabulation had occurred in the instant case; it was, however, a theory that could be an explanation.

At trial Professor Stumpff was asked a lengthy hypothetical question by defense counsel which assumed the acceptance of certain facts relating to the incident, some of which were controverted. At the conclusion of the hypothetical question, defense counsel asked:

Assume further, and lastly, that after the occurrence of the accident somebody says or makes inquiry as to who was driving, and the individual answers, "I was." Do you have an opinion, with a reasonable degree of scientific certainty, based upon your experience and your studies, as to whether or not the facts that I have just given to you and the symptoms that I have just given to you are consistent with confabulation? Yes, or no, please?

(Emphasis supplied.) Professor Stumpff answered "yes."

Neb. Rev. Stat. § 27-702 (Reissue 1979) provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

Expert testimony must meet these requirements in order to be admissible. State v. Ammons, 208 Neb. 812, 305 N.W.2d 812 (1981). At best, it was the opinion of Professor Stumpff that the facts were consistent with confabulation. At no time was there any closer application of the theory to the facts at hand. Whether the facts were consistent with confabulation was not relevant, and his testimony should not have been received. The general rule is:

"Expert testimony should not be received if it appears the witness is not in possession of such facts as will enable him to express a reasonably accurate conclusion as distinguished from a mere guess or conjecture. The witness should not be allowed to express an opinion on an inadequate basis or in respect to facts not disclosed to the jury."

Langfeld v. Department of Roads, 213 Neb. 15, 27-28, 328 N.W.2d 452, 459 (1982). See, also, Fletcher v. State, 216 Neb. 342, 344 N.W.2d 899 (1984).

Where the opinion testimony of an expert witness does not have a sound and reasonable basis, it should be stricken. Danielsen v. Richards Mfg. Co., Inc., 206 Neb. 676, 294 N.W.2d 858 (1980). There was no showing that confabulation occurred or even that the defendant sustained such neurological damage as would support its possible occurrence. As such, there was no basis in fact to support Professor Stumpff's testimony. Where there has been a clear abuse of discretion, the ruling of a trial court in receiving or excluding an expert's opinion will be reversed. State v. Miner, 216 Neb. 309, 343 N.W.2d 899 (1984). The opinion of an expert witness is properly admitted into evidence where the basis of the opinion and the facts on which it is based are before the jury and the opposing party has had an opportunity to cross-examine the expert witness in that regard. Indeed, there is a vast difference between permitting an expert to give an opinion when all the factors necessary to draw a conclusion are in evidence and in permitting an explanation based on assumptions that have no adequate foundation in the evidence. Nickal v. Phinney, 207 Neb. 281, 298 N.W.2d 360 (1980).

The third assignment of error to be considered is the trial court's failure to find that there was not sufficient foundation to permit evidence of the results of alcohol content tests performed on blood and urine samples taken from the body of Larry Priest. Despite plaintiff's objections that a complete chain of custody had not been established, Waltemath was allowed to testify that the alcohol content of his blood was .105 percent and that the alcohol content of his urine was .16 percent.

The record reveals that Dr. Melvin Campbell recalled taking blood samples from both the deceased Larry Priest and Linda Lister. Dr. Campbell said that he took the samples at the mortuary, but he did not remember to whom he passed them for handling. Usually, he said, they are given to the sheriff or mortician.

Sheriff Donald Brown testified that he took blood and urine samples to his office; he did not know where he got the Larry Priest samples, but he did remember who gave him the Linda Lister samples.

Dr. Selliken performed the autopsies of the decedents. He testified that he was given samples labeled with their names from a uniformed individual at the Ainsworth sheriff's office. He transported the samples to his laboratory in North Platte, where he put them in a lockbox used for keeping specimens. Dr. Selliken requested that Waltemath perform alcohol tests on the samples. Waltemath then took the samples out of the lockbox in the laboratory. She testified that one of the samples had "Dr. Selliken's name on it, Brown county, and Dewey Long . . . [and] Priest."

In Raskey v. Hulewicz, 185 Neb. 608, 177 N.W.2d 744 (1970), a challenge was made to the sufficiency of the foundation for the chemical analysis of a urine sample. This court said that "the authenticity of the urine sample must be unequivocally established before its admission into evidence, as an elementary safeguard of a defendant's rights." Id. at 615-16, 177 N.W.2d at 749. In the case at bar there is no evidence as to the origin of the urine sample of Larry Priest. Not only did Dr. Campbell not remember taking any urine samples but the sheriff had no knowledge of where he obtained either the urine or blood samples of Larry Priest. The first time that any urine sample is directly associated with Larry Priest is when Waltemath removed it, along with the ambiguously labeled samples, from the lockbox in her laboratory.

At best, the chain of custody concerning the blood samples is equivocal. Not only did Dr. Campbell not remember to whom he gave the blood samples but the sheriff did not know how he obtained any of the samples from Larry Priest. Once again, Waltemath's testimony is that these blood samples were among those ambiguously labeled samples she retrieved from the lockbox. In State v. Apker, 204 Neb. 577, 578, 284 N.W.2d 14, 16 (1979), this court said:

"We hold that an exhibit is admissible, so far as identity is concerned, when it has been identified as being the same object about which the testimony was given. It must also be shown to the satisfaction of the trial court that no substantial change has taken place in the exhibit so as to render it misleading. As long as the article can be identified it is immaterial in how many or in whose hands it has been." In State v. Langer, 192 Neb. 525, 222 N.W.2d 820 (1974), we said: "Important in such situations are the nature of the exhibit, the circumstances surrounding its preservation and custody, and the likelihood of intermeddlers tampering with the object."

Where an object passes through several hands before being produced in court, it is necessary to establish a complete chain of evidence which traces the possession of the article to the final custodian. If one link in the chain is missing, the object may not be introduced into evidence. State v. Bobo, 198 Neb. 551, 253 N.W.2d 857 (1977).

Appellant also contends that the trial court's jury instructions were improper, specifically, in that no instruction was given on "comparative negligence" and "contributory negligence," and in that no definition of "negligence" was given. Generally, this court will not consider instructions not objected to. Silvey Co., Inc. v. Engel, 204 Neb. 633, 284 N.W.2d 560 (1979). "However, the trial court is under a duty, on its own motion, to correctly instruct on the law and this court may take cognizance of plain error in instructions indicative of a probable miscarriage of justice." Id. at 636, 284 N.W.2d at 562.

While these omissions, standing alone, might not warrant reversal, prejudicial error is evidenced when those omissions are coupled with the factor that the jury resorted to a dictionary to supply itself with certain definitions. In fact, the final error assigned in this case concerns the jury's use of a dictionary during its deliberations. The affidavits of two of the jurors indicate that the bailiff provided the jury with a standard English dictionary, from which definitions for "negligence," "contributory," and "preponderance" were obtained. One of the jurors stated that the dictionary was requested because some of the jurors were having a dispute regarding the burden of proof on certain elements of the case. He further stated that "the use of the dictionary in the jury room with the issue of the burden of proof and the word `preponderance' had an impact of [sic] the decision of the jury."

In Schreiner v. State, 155 Neb. 894, 54 N.W.2d 224 (1952), this court considered an assignment of error regarding a jury's use of a dictionary. There, the jury had the dictionary for only a few minutes, and the foreman advised the trial court that they had looked up a definition but "`we didn't get too much out of it.'" Id. at 897, 54 N.W.2d at 226. The remaining jurors advised the trial court that the reference to the dictionary had not affected their deliberations. Under those circumstances no error was found. However, in reaching its decision, this court reviewed the standard set out in previous cases:

In Matters v. State, 120 Neb. 404, 232 N.W. 781, a juror during deliberations produced a dictionary and read from it the definition of some words which were contained in an instruction. We held: "This conduct was highly improper. The jury should have relied solely upon the evidence for the facts, and upon the court's instructions for the law, of the case. Not every violation of the proprieties, however, is sufficient to reverse a judgment. It is only such misconduct of the jury as is calculated to prejudice the substantial rights of the defendant that is ground for a new trial. . . ."

Id. at 898, 54 N.W.2d at 226.

In sum, it cannot be said that the combined failure of the trial court to properly instruct the jury and the jury's resort to a dictionary to fill in the blanks did not result in prejudice to the plaintiff.

For the reasons stated the judgment is reversed and the cause remanded for a new trial.

REVERSED AND REMANDED FOR A NEW TRIAL.