ALLEN, Acting Chief Judge.

This appeal by the defendant-appellant is from a final judgment entered on a $45,000 jury verdict in appellee's favor.

The plaintiff-appellee in this case, Charles Cochran, is an automobile mechanic. He has been engaged in this occupation approximately twenty years and was so employed by Bartow Ford Company on March 19, 1963. On the aforementioned date the plaintiff, pursuant to a bona fide work order, had undertaken the repair of a 1962 Ford generator which entailed replacement of the rear bearing. Plaintiff obtained a new bearing manufactured by the defendant, Ford Motor Company, from Bartow Ford Company's parts department, but during the actual repair procedure the new bearing shattered. A portion of the bearing struck plaintiff in the left eye inflicting such a severe injury that the eye had to be removed. The plaintiff sued Ford Motor Company asserting that the new bearing contained a manufacturing defect, which had caused it to be unfit for its intended use; and, that the defendant had committed a breach of warranty.

The defendant moved to dismiss the complaint due to the plaintiff's failure to allege the nature of the bearing's defect. Upon denial of the above motion, the defendant sought an order permitting it to depose plaintiff's expert witnesses in order to determine what plaintiff's contentions would be regarding the alleged defect. The plaintiff had retained the services of two metallurgical experts to perform certain tests, evaluate the results, arrive at logical conclusions and render opinions concerning the nature of the bearing's defect. The trial court ruled that plaintiff's expert witnesses could not be deposed as to opinions or conclusions found by them while attempting to learn the nature of the defect.

We think the trial judge ruled correctly and affirm the judgment of the lower court.

Ford Motor Company v. Havee, Fla.App. 1960, 123 So.2d 572, is a case not so unlike the one presently before this court. The Havee case was a personal injury action against an automobile manufacturer, Ford Motor Company. An order had been entered by the circuit court requiring the manufacturer to produce a report given to it by the expert who examined the defective tie rod of the automobile in which the plaintiffs were injured. The appellate court held that the plaintiff was not entitled to a copy of the expert's report of the examination of the tie rod where good cause had not been shown. The conclusion reached by the court in the Havee case, supra, seems to be buttressed by the fact that the circuit court had failed to designate experts to examine the tie rod or prescribe the condition and scope of the examination. And further, when the tie rod was returned undamaged and not disassembled or changed, the information contained in the report could have been discovered by plaintiffs performing their own examination of the tie rod.

In the Havee case, the court said:

"It is a well-settled principle of our adversary procedure that discovery may not be used to secure private memoranda of an adverse party's counsel formed in the course of his legal duties. Hickman v. Taylor, 329 U.S. 495, 67 S.Ct. 385, 91 L.Ed. 451. This case was followed by the Supreme Court of Florida in Atlantic Coast Line R. Co. v. Allen, Fla. 1949, 40 So.2d 115, which held that discovery may not be used to examine the `work product' of the adverse party and it matters not whether the product is the creature of the party, his agent or his attorney." (Emphasis supplied.)

The court then recognized that there might be certain exceptions to the above rule and listed the following circumstances in which discovery would be permitted:

"`* * * (1) when witnesses are no longer available or can be reached only by great difficulty, (2) if it gives clues to relevant facts that cannot be secured otherwise, or (3) for purposes of impeachment.'"

In the case sub judice the Ford Motor Company is on the opposite side of the fence from what it was in the Havee case. Here, Ford is seeking to depose plaintiff's experts in order to discover what they would say with regard to the alleged defect in the bearing. We recognize that in the instant case the defendant wanted to secure oral testimony from plaintiff's experts in contrast to the Havee case where a copy of a report was sought. For us to say that findings by experts during the performance of examinations and tests in a particular case are not subject to discovery proceedings if reduced to writing, and to hold otherwise if the findings have not been reduced to writing, would be an absurdity in which this court does not care to indulge. Opinions and conclusions lodged in the minds of the plaintiff's experts are no more subject to discovery proceedings than a detailed written report would be unless, of course, the report comes within one of the exceptions to the rule which we deem to be inapplicable in the instant case.

This court has recognized that an expert witness is not subject to discovery with regard to expert testimony in the recent case of Hartstone Concrete Products Company, Inc. v. Ivancevich, Fla.App. 1967, 200 So.2d 234. In that case the defendant attempted to call a physicist during trial to impeach the plaintiff's traffic expert. The trial judge refused to permit the physicist to be called because he was not listed on the pretrial order. We reversed and pointed out that even though the plaintiff's expert witnesses were listed in the pretrial order, the defendant did not have the right to discover their opinions regarding the accident. Since the defendant could not have discovered the testimony of the plaintiff's expert witnesses, the defendant would have no way of knowing until trial what the testimony of the expert was to be, and prior to that time would not be able to obtain any type of impeachment evidence. The court further stated that the plaintiff could not have been prejudiced by the defendant's failure to list the physicist in the pretrial order, since the plaintiff could not have ascertained the physicist's testimony by discovery anyway.

Appellant's contention that no specific defect in the generator bearing was proved is untenable. Arguments based on this same theory were advanced and rejected in the "grinding wheel" cases of Trowbridge v. Abrasive Co. of Philadelphia, 3 Cir., 190 F.2d 825, and Norton Company v. Harrelson, 1965, 278 Ala. 85, 176 So.2d 18.

In the Trowbridge case, supra, the Court said:

"Defendant argues on appeal that there was no evidence of negligence because no witness identified any defect in the wheel. Defendant's version of the law would cast upon a plaintiff in a case such as this an almost impossible burden; for where an abrasive wheel disintegrates, direct proof of the existence of a defect might be virtually impossible. See Zesch v. Abrasive Co. of Philadelphia, 1944, 353 Mo. 558, 183 S.W.2d 140, 156 A.L.R. 469. The burden of the scientific limitations of our society should not be cast on injured plaintiffs in circumstances such as existed in the instant case." (190 F.2d p. 828)

We agree. In the case sub judice, as in Trowbridge, the plaintiff did not and could not show a specific defect in the bearing which injured him. It was impossible to do so. Plaintiff's experts performed at least four different tests on similar bearings and concluded from the tests that the bearing involved in the accident ruptured because of "a metallurgical defect." We think that the plaintiff did all that could have been reasonably expected of him under the circumstances.

Appellant next contends that the court erred in allowing the jury to consider the testimony and opinions of plaintiff's experts. We find no error in the court's handling of this matter. The testimony and opinions of plaintiff's experts were based on the tests which they had performed.

As the Alabama Supreme Court said in the Norton case, supra, (176 So.2d p. 22):

"* * * In this case * * * the appellant addresses his argument to the wrong forum. Admittedly not all evidence of experiments or demonstrations is admissible. However, we have said:

"`Of course, there must be similarity of conditions to give an experiment sufficient probative value to warrant its admission, and if the conditions were dissimilar in an essential particular, the evidence should be rejected. But the authorities are to the effect that it is not necessary that the conditions should be exactly identical. A reasonable or substantial similarity suffices, and lack of exact identity affects only the weight and not the competency of the evidence. It is for the court to determine whether the conditions are sufficiently similar to warrant admission of this proof, and much must be left to the sound discretion of the trial judge.' Louisville Nashville R. Co. v. Sullivan, 244 Ala. 485, 13 So.2d 877."

The tests performed by plaintiff's experts satisfied the trial judge. The record fails to show that the judge abused his discretion in allowing the results of the tests to be heard by the jury. Absent such a showing of abuse, we find no error.

We have carefully considered the additional arguments and issues propounded by appellant and find them to be without merit.

The judgment of the lower court is, therefore, affirmed.

Affirmed.

SHANNON and HOBSON, JJ., concur.