In July, 1946, Reynolds Metals Company, a Delaware corporation, acquired through lease, an aluminum plant belonging to the Government and located at Troutdale, Oregon. It commenced operation of its first potline for the production of aluminum in September of that year. In the operation of the plant chem *323 ical compounds containing aluminum were collected in the reduction cells of the so-called “pots” and were reduced or separated by the process of electrolysis or passage of current through the cell. In the process, temperatures up to 1775° F. were developed. As the compounds with which the cells were charged, (cryolite or sodium fluoride or calcium fluoride and aluminum fluoride) contain large percentages of fluorine ranging around 50 percent, a considerable portion of fluoride material was volatilized in the process and reached the atmosphere. 1

Shortly after the operation of the plan began, and in December, 1946, the appellees in these three appeals, Paul Martin, his wife, Verla Martin, and his daughter, Paula Martin (now Mrs. Yturbide), moved to their cattle farm or ranch near Troutdale located about a mile to a mile and one-half from the aluminum plant, and they resided there until November, 1950. The law suits out of which these appeals grew were based upon claims that during the period of their residence on the farm they were poisoned by fluorides which originated at the appellant’s plant and were borne on the air to the farm where they breathed these fluoride effluents claimed to be highly toxic and also ate vegetables growing in their garden which also had been absorbing the same toxic elements. 2

The principal question presented here is whether the evidence adduced at the trial below, (the three cases were tried together and upon the same evidence) was sufficient to permit the case to go to the jury. Upon the part of the appellant it is contended, (1) that there was insufficient proof to demonstrate that the damage to the persons of the plaintiffs, of which they complain, was caused by the fluorides escaping from defendant’s plant; and (2), that there was no evidence of any negligence, or any other breach of duty, on the part of the defendant.

With respect to the question of causation: — whether the escaping fluorides did in fact cause plaintiffs’ injuries, —the evidence was sufficient to warrant the jury’s conclusion that the escaping fluorides were the cause of the injuries. It was not disputed, in fact it was stipulated, that “fluorides in some quantities and forms, did escape, when the potlines of said plant were operating, from defendant’s plant.” The pre-trial stipulation, after stating the specific fluoride combinations which did escape in the form of gases, liquids and solids, and which became air-borne, recited that “portions thereof have settled at various times upon the lands” of the Martins.

A horticulturist from the Oregon State College, in the years 1948, 1949 and 1950 made test samples of plants grown on selected plots in the vicinity of the Martin property, for the purpose of determining the fluorine content of such plants. Some of the test plots used were on the Martin property and as close as 1.1 miles from the aluminum plant. Others located in the same direction from the plant were farther away. Those plants tested on plots nearest the aluminum plant showed substantial amounts of fluorine content. The plants tested on plots which were farther away from the plant showed a subtantially decreasing fluorine content, thus indicating that relatively speaking the nearer the aluminum plant the greater the concentration of the gases, fumes and particulates, (i. e. fine solids). In making such test, plants *324 which readily absorbed fluorine compounds such as buckwheat and gladioli were used so as to facilitate the testing. The results therefore did not indicate the quantities of fluorine which would likely be found in garden vegetables such as those the Martins grew in their garden plot for their domestic use.

In general there was an absence of proof as to just what quantities of fluorides contained in these gases, fumes, and particulates, passed over the Martin land or were inhaled by them or ingested from the garden vegetables eaten by them. There was, however, proof that these fluorides were toxic. One of the chemists who testified stated without challenge that hydrofluoride acid or hydrogen fluoride, one of the effluents from the plant, was “quite toxic” and that “the books on chemistry warned against inhaling it.” Appellant concedes that such fluorides “are poisonous in excessive amounts.”

That very large quantities of these gases, fumes, and particulates, did leave the plant and were diffused into the air, was unquestioned. Appellant's own exhibits disclose that with the equipment which was used to control the escape of gases during the period that the Martins lived on their farm, hundreds of pounds of these effluents escaped each day. Thus in the year 1947, the amount of fluorine alone escaping per day from the plant averaged 2845 pounds. Comparable amounts escaped in the years 1948-1949 and throughout the first half of 1950, after which time the appellant began the installation of a new control system which was much more efficient in arresting the escape and fall-out.

There is no showing as to where these large quantities of effluents finally settled. The experiments mentioned above would indicate that the greater portion settled on those areas nearest the plant, and those areas included the Martin farm. That the effluents did have some degree of toxic or harmful effect was indicated by proof that cattle upon the Martin place shewed damage from fluorosis. Generally with respect to cattle it does not appear to have been controverted that cattle damage from an aluminum plant is a fairly common phenomenon. 3

Of course it does not follow from mere proof of some damage to cattle on the Martin place that the plaintiffs’ physical injuries were due to excessive amounts of fluorides from the plant. Cattle get their whole food from grass and ingest large quantities. A human diet contains a relatively small proportion of vegetables. For the purpose of this opinion we assume that the evidence that cattle on the Martin place showed injury to health from fluorides has no significance here other than to show the fluorides did reach the Martin place in some quantity and that in some quantity the ingestion of affected plants may possibly cause illness in a mammal. 4

*325 A significant bit of testimony adduced was proof that glass in the Martin home became etched by acid, probably hydrofluoric acid which was one of the effluents from the aluminum plant. One of the expert witnesses, the British doctor who had some prior experience with similar etching of glass located near industrial plants abroad, testified that the glass from the Martin window which he was shown during the testimony was an indication of excessive quantities of fluoride contamination in the atmosphere.

Although the cases of the Martins are unique in that they were unable to produce either from medical literature or expert witnesses histories of persons situated as they were, namely, persons not working in the plant but simply living outside and near the plant, who developed symptoms similar to theirs in consequence of the fall-out of emanations from an aluminum plant, (see footnote 5a infra) nevertheless they did produce substantial medical testimony which did connect their disabilities and physical injuries with the fluorides escaping from the plant. Of course there was also testimony to the contrary, but the evidence adduced on their behalf, principally from the British medical expert referred to in footnote 3, supra, and from a Dr. Capps, a Chicago specialist on diseases of the liver, was substantial, and in our view, fully worthy of credit. Dr. Capps had been consulted by the Martins in November, 1951; Dr. Hunter, the British expert, examined the Martins shortly before the time of the trial; he also reviewed the laboratory and other tests made with respect to them by Dr. Capps and by other physicians who had examined them. Their’s was not the only medical testimony adduced in support of the claims of the Martins, but it was probably the most complete. Their qualifications to testify was not only adequate but their experience with the subject upon which they testified was outstanding. 5 For the purposes of this opinion it is sufficient to say that the evidence clearly warranted the jury in accepting the testimony and conclusions of these doctors. Hence there is no point in undertaking to detail their testimony at length. 6

*326 The medical witnesses expressed confidence in their conclusions because of the fact that they found substantially the same symptoms in the three individuals, thus lessening the possibility that the symptoms were attributable to individual idiosyncracies. Neither one was able to testify as to the degree of concentration, — how many parts per million of fluorides, — would be required to bring about the injuries found, but both attached some significance to the fact that the window in the house was etched by acid. As stated by Dr. Capps, “I think that if there is enough fluorine to etch a window, it should be able to etch a lung.” Dr. Hunter related his own experience in finding in England a family with the same symptoms as the Martins who lived in a house near an industrial plant 7 whose windows were similarly etched and said that the etching he observed in the glass taken from the Martin home was an indication of excessive quantities of fluorine in the atmosphere. He referred to studies made in 1946 warning the industrial world “against throwing into the atmosphere an effluent which would etch glass.”

A further circumstance which the medical witnesses emphasized was that in the case of the Martins their abnormal symptoms were reduced and lessened after they had moved away from the vicinity of the plant. We are thus led to the conclusion that the jury was warranted in finding, as they did, that physical injuries to the Martins were caused by the fluoride emanations from the plant which found their way to the Martin place.

On behalf of the defendant evidence was adduced to show that only a very small concentration of fluorides could have reached the Martin home. We think the jury were warranted in accepting plaintiffs’ testimony, and the inferences therefrom, that excessive amounts of fluorides reached them, for they actually suffered fluoride poisoning. There was no proof on the part of plaintiffs as to what particular percentage or concentration of fluorides was necessary to produce such results. Plaintiffs were obliged to accept in that respect the concession made by the defendant to the effect that such fluorides are “poisonous in excessive amounts.” The only possible conclusion is that because the damage was actually caused by the fluorides, they somehow managed to get there in excessive quantities. We think we must accept this as a permissible finding of the jury.

The important question is whether there was proof of any negligence or other breach of duty on the part of the defendant which brought about these injuries as its proximate result.

In our former opinion, in the case of Paula Martin, (Reynolds Metals Company v. Paula Martin Yturbide, April 24, 1957, not reported), this court expressed the view that these plaintiffs were entitled to recover under the doctrine of Fletcher v. Rylands, 3 H. & C. 774, (affirmed in Rylands v. Fletcher, L-R. 3 H.L. 330), and referred to certain *327 Oregon decisions which were thought to recognize a doctrine of strict liability, or liability without fault. What we said there about that doctrine and its possible application here is now withdrawn. As is hereafter developed, a discussion of that problem is unnecessary to a decision of this case. Furthermore, this case was not submitted to the jury which heard it upon any such theory. 8

Appellant argues that there is a complete absence of proof of any want of care on its part in the operation of the aluminum plant. In this respect its contention is that before the plant was put in operation it equipped it with facilities for minimizing the amount of effluents that escaped from the plant. Each potroom building was equipped with a ventilator extending the full length of the building at the roof or ceiling. Air was supplied to the room through louvre doors at the base and the heated air carrying the escaping gas, fumes and particulates moved upward to the ventilators. At the ventilators a washing system was installed which consisted of a series of sprays running the full length of the building so that any air that left the building had to pass through these sprays. After the spray system was first started, baffles were inserted to intercept and condense the moisture which came from the spray into the escaping air; in other words, to take “mist” out of the air. The result of this spray system was that it removed some 60 percent of the fluorides from the air coming through the roof so that only 40 percent of these compounds actually escaped from the plant.

Defendant’s Assistant Vice-President testified that during this period from 1946 to 1950, when the spray system of fluorides elimination was in use, the company was carrying on experiments to devise and find better ways of making a more efficient capture of the effluents. In 1948, a pilot plant was designed for the purpose of trying out an improved system of elimination, and as a result of the experiments with that pilot plant, it was arranged about May, 1949, to change over to a new system. The installation of that was completed in November, 1950. The new system involved the placing of hoods over each pot and the installation of fans which sucked the air coming from the pots upward and forcing it through sprays and wash-towers. The result of the use of this improved system was to eliminate 90 percent of the fluoride compounds from the air which escaped from the plant. While, as we previously indicated, the average pounds of fluorine escaping per day during 1947 were 2845 and in the month of March, 1950, they were as high as 3988 pounds, yet beginning in July, 1950, when the new system began to be used, there was a sharp reduction so that in November and December, 1950, after the completion of the improved system, the average daily flow of fluorine escaping was reduced to 643 pounds. In short, by this improved method, the portion of the emanating fluorides which reached the air was 10 percent of the total as against 40 percent under the earlier system.

In determining the duty of care the law imposed upon the defendant, it is to be borne in mind that the ordinary care which the law requires of such a defendant is measured in part by the defendant’s knowledge of potential dangers. Thus a seller of fur coats who actually knows that some few persons are especially susceptible to poisoning by a particular fur dye must take precautions accordingly, and greater precautions than would be required if he were ignorant of this particular susceptibility. *328 Gerkin v. Brown & Sehler Co., 177 Mich. 45 , 143 N.W. 48 , 48 L.R.A.,N.S., 224. As stated in the Restatement of Torts, § 289, in determining the standard of reasonable conduct, “the actor should recognize that his conduct involves a risk of causing an invasion of another’s interest, if a person, possessing * * * such superior perception as the actor himself has * * * would infer that the act creates an appreciable chance of causing such invasion.” Comment (n) under this section headed “Superior knowledge” recites: “If the actor has a wider knowledge, whether gained from personal experience or otherwise, the propriety of his judgment as to the risk involved in his conduct is determined by what a reasonable man having such knowledge would regard as probable.” This requirement of conduct proportioned to the superior knowledge and intelligence of the actor is developed at length in discussions on the subject in Prosser on Torts, 2nd ed., pp. 132 to 134, and in Harper & James, The Law of Torts, §§ 16.5 and 28.4. Not only that, but it was the duty of one in the position of the defendant to know of the dangers incident to the aluminum reduction process. As the last named authors point out in § 16.5, “The manufacturer must learn of dangers that lurk in his processes and his products.”

The record here amply demonstrates that the defendant was fully aware of the potential dangers in the escape of fluorides. It is not questioned that numerous claims of injury to cattle from the operation of such plants had been presented to this and other manufacturers. The defendant’s senior chemist at the Troutdale laboratory testified that he was familiar with the literature upon the toxic qualities of these fluorides; that he knew that in quantities they were toxic; that some of these were heavier than air and would reach the earth after being air-borne; that quantities of them would drift in the direction of the Martin land and that there were publications which dealt with dangers to human beings from fluorides.

The question is whether in view of the knowledge which the company was reasonably required to have as a manufacturer in this field, and the knowledge which it actually did have of the potential dangers of these escaping fluorides there was evidence of a failure to use reasonable care for the protection of persons in the situation of the Martins.

On behalf of the defendant it was testified that at all these times defendant was utilizing and had installed the latest and best known means of protecting against the escape of the fluorides; that the company, continually studying the problem, installed an experimental pilot plant to develop a new process and then changed over to the improved process as soon as its feasibility was developed.

On behalf of plaintiffs it was urged that the evidence warranted a conclusion that the company did not exercise the care in this direction which it might have done in that it allowed the plant to be operated with the less efficient elimination system from 1946 until 1950 whereas proper care required an earlier use of the better devices.

The only bit of evidence which was taken to sustain such a contention is found in the testimony of Dr. Hunter who testified as follows: “Q. (By Mr. Mead): Doctor, are you familiar with an induced air system of dust control or fume control? A. Naturally, I haven’t been trained as a ventilating engineer. But since I first studied industrial hygiene in the Harvard School of Public Health with the famous Phillip Drinker —he is a ventilating engineer — and because of the work which I have to do and so teach, which is industrial medicine, I am, naturally, interested in systems of ventilation. And in all of the chapters of that book there is an exact description of systems of ventilation which are used in various parts of the world. And in 1946 it was well-known to all industrialists the world over that exhaust ventilation could be effectively applied to remove these effluents.”

Possibly we can assume that the question relating to an “induced air system” *329 referred to the sort of system established by the company in 1950. Testimony that such a system was well known in 1946 would permit an inference that failure to use that better system at that time constituted negligence. We believe it is unnecessary in reaching our ultimate conclusion here to draw such inference from this single answer in this long record for we are of the opinion that the trial court properly submitted to the jury the question of negligence on the part of the defendant on the theory that this was a case correctly permitting the application of the doctrine of res ipsa loquitur.

The court instructed the jury as follows: “Under the law and the facts of these actions, the defendant was the sole operator and in exclusive possession and direct control of the aluminum plant involved during the time involved, namely, between on or about September 23, 1946, and November 30, 1950. Further, that under the ordinary course of events, it is unexpected that persons being in the vicinity of such a plant would be injured or harmed by fluorine compounds emanating therefrom, and that such a mishap would not occur.” The statement in the first of these two sentences is unquestionable; the second sentence is in conformity with defendant’s theory and much of the evidence adduced by it at the trial. The instruction was consistent with the testimony that the plant as equipped and operated, would not cause fluorosis to persons who had not been employed in the plant itself. The court noted that there was no evidence as to the amount or concentration of fluorides that would bring about fluorosis in the case of such persons. While during the critical period here 40 percent of the gases, fumes and particulates got by the sprinkler system and escaped into the air, the fluorine contents aggregating hundreds of pounds all told, the tests made at various localities near the plant and on every side thereof, according to defendant’s witnesses, produced no significant concentration of fluorides at the time the tests were made. Hence, the court, accepting this proof, told the jury that, “under the ordinary course of events, it is unexpected that persons being in the vicinity of such a plant would be injured or harmed by fluorine compounds emanating therefrom.”

In so ruling the court in effect determined that the accident was of a kind which ordinarily does not occur in the absence of someone’s negligence. That is a prime condition for the application of the so-called rule of res ipsa loquitur. Said the court (135 F.Supp. at page 381): “Here was an instrumentality that was in the exclusive possession and control of the defendant. Something that ordinarily we would not expect to happen did happen and damage resulted. This is the pure and simple test as the Court sees it, of res ipsa loquitur.” 9 The court was using language similar to that of Ritchie v. Thomas, 190 Or. 95 , 224 P.2d 543 , 551: “An inference of negligence may arise only when injury is caused by an instrumentality which is under the control and management of the defendant, and when the accident is such as, in the ordinary course of events does not happen, if those who have the management use ordinary care.” Such is the test generally approved in such cases. See Prosser, Law of Torts, 2nd Ed., p. 201; Harper and James, The Law of Torts, § 19.6.

Just what caused this not-ordinarily-to-be-expected scattering of fluorides in “excessive amounts”, was not specified. That they were so scattered *330 the jury was warranted in finding. What concerns us here is that the trial judge was justified in paraphrasing the language of the Suko case, (supra, note 8): “In the ordinary course of matters the mishap would not have occurred except through carelessness. * * * ” Whether this want of care was in faulty construction, faulty inspection, or faulty operation of the fluorides elimination system, plaintiffs were not required to show, for wherever the fault lay, it was necessarily that of the defendant which had exclusive control. Of course the inference of negligence referred to in Ritchie v. Thomas, supra, is merely a permissible one, Bedal v. Hallack and Howard Lumber Company, 9 Cir., 226 F.2d 526, 538 , but it carries the case to the jury.

Viewing plaintiffs’ case in the most favorable light, as we are required to do, we have the following: (1), Damage to persons in the position of the Martins was possible from “excessive amounts” of fluorides; (2), the Martins were poisoned and from this plant and by these fluorides; (3), from this it is permissible to infer that “excessive amounts” reached them, — that there were such excessive amounts; (4), the emission of excessive amounts from the plant was circumstantial evidence of negligence. Res ipsa loquitur.

Appellant says that the res ipsa rule was misapplied here. It quotes from this court’s decision in Orr v. Southern Pacific Company, 9 Cir., 226 F.2d 841, 843 , the statement: “Res ipsa loquitur is a procedural rule as to evidence. It is never valid in the face of facts which adequately explain the occurrence.” Here, it asserts, it made an adequate explanation of the occurrence. It says: “Reynolds, as has been shown, used the greatest possible care, far above that required by any reasonable risk of human injury.”

The Orr case presents a different situation for the judge who found the facts decided for the defendant. In the present case the jury might have accepted defendant’s proof and explanation but obviously it did not. It was not obliged to do so. The procedural problem which faces any defendant such as this, as it has developed in the cases, is discussed at some length by both Prosser and Harper & James. The latter say, (§ 19.12): “Defendant in a res ipsa case may show by evidence how the accident happened, but ordinarily this does not entitle him to a directed verdict. The jury may still reject the evidence or find that the explanation does not preclude the likelihood of negligence. * * * In most res ipsa loquitur eases defendant cannot definitely explain the accident. His usual defense consists of proof of the precautions he did take in constructing, maintaining, and operating the injuring instrumentality. Once in a great while such proof may conclusively show that the accident did not in fact happen, or that it was not caused by defendant. But this is seldom the outcome. And where it is not, defendant is in this dilemma: the less effective his precautions to prevent the occurrence the more apt they are to appear negligent; the more effective the precautions testified to, the less likely they are to have [been] taken in this case since the accident did happen.” 10

*331 Here the jury were instructed at length to the effect that the plaintiffs’ claims were based upon their assertion that defendant had been guilty of negligence. The nature of their claim and the negligence asserted was stated by the court to the jury in substantially the same language as that set forth in the pretrial statement quoted above. The court stated the contentions of the defendant in respect to this claim of negligence; it defined negligence and set forth that the plaintiffs were required to prove not only negligence but that such negligence was the proximate cause of the injury or damage. There is no contention here that these terms were not adequately defined in the instructions. 11

When res ipsa loquitur is applied the facts of the occurrence warrant the inference of negligence. They furnish substantial evidence of negligence where the direct evidence of it may be lacking, but they make a case to be decided by the jury. 12 In finding itself in the dilemma mentioned by Harper & James in the language quoted above, appellant is in no different situation than any other defendant who finds itself in a res ipsa case and before a jury. It is not, as appellant asserts, the victim of the application of a rule of absolute liability.

Appellant contends that the court misconstrued and misapplied the res ipsa loquitur doctrine in that it failed to use the language quoted above from Ritchie v. Thomas to the effect that the-inference of negligence permissible under that rule may arise “when the accident is such as, in the ordinary course of events does not happen, if those who have the management use ordinary care.” (Italics ours.)

We have noted above that it is plain that the trial court had this particular requirement in mind when it stated in its instruction to the jury that “under the ordinary course of events it is unexpected that persons being in the vicinity of such a plant would be injured or harmed by fluorine compounds emanating therefrom.” The court’s reported opinion, previously cited, dealing with the *332 rule of res ipsa loquitur, in citing and relying upon Suko v. Northwest Ice & Cold Storage Co., supra, which fully and correctly defined the fact situations which warrant res ipsa loquitur, was not confused as to what the requirements for the application of that rule were. That the language used by the court in its instruction to the jury was not in the precise language employed in Ritchie v. Thomas, supra, is not a circumstance of which appellant may here complain. In the first place the instructions were not objected to upon any such ground. 13

Appellant objected first that the doctrine of res ipsa loquitur was not applicable in a ease of this kind; and second, that it was not applicable here since any inference of negligence had been rebutted by affirmative proof. The point now sought to be made — that the definition of the conditions under which the rule would apply were incorrectly stated — were not the subject of the objection. See Rule 51, F.R.Civ.P., 28 U.S.C.A. Furthermore, any inaccuracy in the court’s reference to the res ipsa loquitur rule as explained to the jury was of no consequence since under the Oregon rule the court was not required to instruct the jury on this subject. Ritchie v. Thomas, supra. We have no occasion to consider whether it was error to refer to the rule of res ipsa loquitur in instructing the jury for no objection or exception was taken on that ground. 14 We cannot see that the instruction was prejudicial, — certainly not in any respect referred to in the appellant’s objections or exceptions to the charge.

Appellant contends that the claims of plaintiffs Paul and Verla Martin are barred by the Oregon two year statute of limitations, Oregon Rev.Stat. § 12.110. These suits were filed January 7, 1952. 15 The question is whether the causes of action arose prior to January 7, 1950. The period of the Martins’ exposure to the fluorides continued from December, 1946, to November, 1950. There was testimony that the symptoms had appeared before the end of 1948. Appellant’s argument is based upon the case of Piukkula v. Pillsbury Astoria Flouring Mills Co., 150 Or. 304 , 42 P.2d 921 , 44 P.2d 162 , 99 A.L.R. 244 . In that case the plaintiff suffered an occupational disease caused by his exposure to the flour dust in the mill. On September 17, 1930, plaintiff experienced difficulty in breathing; in January, 1931, he quit his work in the mill; on January 22, 1933, he, for the first time, realized the extent and seriousness of his illness, and on June 8, 1933, he instituted an action to recover damages for the injuries to his health. The action was held to be barred by the two year statute.

We think that that case is distinguishable from the present one, which would appear to be covered by the rule laid down in the later cases of Shives v. Chamberlain, 168 Or. 676 , 126 P.2d 28 , and Hotelling v. Walther, 169 Or. 559 , 130 P.2d 944 , 144 A.L.R. 205 . Those were cases in which the plaintiff sought recovery for medical malpractice because *333 of the alleged negligent treatment of the patient. In each case the treatment continued for a long period of time. It was held that the statute did not begin to run until the treatment was discontinued. In the first of these cases the court said: "This continued treatment, * * * constituted a continuing tort causing the .statute of limitations to start only when such treatment ceased.” 126 P.2d 31 . In the second case it was said: “The alleged negligent treatment of the plaintiff must be considered as a whole. Plaintiff was not obliged to split his cause of action. The continued negligent treatment constituted but a single cause of action. Where the tort is continuing, the right of action is continuing. * * * The rule that the statute runs from the last date of the continuous negligent treatment is just and equitable.” 130 P.2d 946 .

In the Piukkula case, on which appellant relies, the period of the receipt of damages from dust ended when the plaintiff quit his job and left the harmful place. This was more than two years before the commencement of the suit. In the case now before us the tort, which was a continuing one, had not terminated in November, 1950, and the statute had not run when the suit was filed in January, 1952.

Appellant also assigns error in the cases of Paul and Verla Martin in the action of the court in limiting the examination upon deposition of two doctors as to information given to these doctors by the two plaintiffs when they were consulted for treatment. Shortly after the actions were commenced, the appellant served upon each of these plaintiffs written interrogatories pursuant to Rule 33 of the F.R.Civ.P. and thereafter appellant took their depositions pursuant to Rule 26 of the same rules. From these it appeared that Paul and Verla Martin had consulted a Dr. Hills and a Dr. Proctor in connection with their claimed injuries. Shortly before the trial appellant sought to take the depositions of these two doctors and plaintiffs moved under Rule 30(b) to have the scope of examination limited so as not to permit examination of either doctor concerning information acquired by them in attending the plaintiffs and which was necessary to enable the doctors to prescribe or act for them. Appellant opposed the motion on the ground that the Martins had waived their privilege with respect to such information by disclosing on their interrogatories and in their depositions facts concerning their physical condition. Holding that the answers of the plaintiffs in their interrogatories and statements on the deposition did not reveal either an express or implied waiver of their privilege created by the Oregon statute, the court granted the order requested limiting the extent of the inquiry in the taking of the depositions.

In dealing with this assignment of error in the opinion of a division of this court on April 24, 1957, (unreported), this court said: “The federal courts look to the state law to determine whether the physicians bear a confidential relationship to their patients. Engl v. Aetna Life Insurance Co., 2 Cir., 1943, 139 F.2d 469 . The applicable Oregon statutes provide: Or.Rev.Stat. § 44.040 (1) (d) : ‘A regular physician or surgeon shall not, without the consent of his patient, be examined in a civil action, suit or proceeding, as to any information acquired in attending the patient, which was necessary to enable him to prescribe or act for the patient.’ Or.Rev.Stat. § 44.040(2): ‘If a party to the action, suit or proceeding offers himself as a witness, it is deemed a consent to the examination also of a physician or surgeon on the same subject/

“Appellant contends that Verla Martin had ‘offered’ herself as a witness because when made a witness by appellant by interrogatories submitted under F.R. C.P. 33 and questioned in a deposition taken by appellant under F.R.C.P. 26 in which her attorney sought no cross-examination, she had not availed herself of her privilege to refuse to answer questions concerning her treatment by two Maryland doctors but had testified concerning their treatment and advice.

*334 “Appellant’s interrogatories were answered in March, 1952 and its questions in the deposition in May, 1952. Instead of attempting to take the depositions of these doctors promptly, it was not until two and a half years later when the case was set for trial on July 25, 1955, that appellant gave notice of taking of the doctors’ testimony in Baltimore on July 16, 1955. Because of the expense to the Martins in sending a lawyer to Baltimore, Mrs. Martin moved for a ruling on the right to take her physicians’ testimony and the district court ruled as follows:

We think for the reasons thus stated it must be said that at the time of the ruling complained of (made on August 6, 1955), prior to the commencement of the trial (which began August 25, 1955), the court was not in error in holding that at the time the proposed depositions were noticed these plaintiffs had not offered themselves as witnesses within the meaning of the Oregon statute. In other words, the ruling of the trial court was correct when made. At the time of the trial these witnesses did offer themselves as witnesses and at that time waived their privilege. At that time Drs. Hills and Proctor would have been proper witnesses. They were not then called, nor was the court at any time during the three weeks’ trial asked for a continuance or an order modifying its earlier limitation to permit at the time an enlarged inquiry on deposition to be made of Drs. Hills and Proctor. Before the court could be put in error in respect to the taking of the testimony of these doctors, defendant must have asked to take the testimony after the privilege had been waived. We hold that in the circumstances appellant has failed to demonstrate that the action of the court in respect to these depositions was prejudicial.

*335 Cross-Appeals Respecting Costs

Each of the three plaintiffs filed cost bills. In the Paul Martin and the Verla Martin cases defendant filed objections to the amounts claimed for witness fees on the ground that the identical items had been claimed, and were allowed in the Yturbide case, asserting that multiple witness fees could not be recovered, since the three actions had been consolidated for trial and the witnesses had testified but once. The court sustained the objections and plaintiffs cross-appealed.

Although an earlier decision of this court, National Union Fire Ins. Co. of Pittsburgh, Pa. v. California Cotton Credit Corp., 76 F.2d 279, 289 , approved allowance of double costs, the statute on which that result was based has been superseded by Rule 42(a) which authorizes consolidation of actions such as these “to avoid unnecessary costs or delay.” We think the better rule is that stated in Copeman Laboratories Co. v. Norge Division of Borg-Warner Corp., D.C., 89 F. Supp. 161, 163 , which suggests that doubling the witness fee allowance where there has been a consolidated trial would operate to frustrate the stated purpose of Rule 42(a) “to avoid unnecessary costs.”

Plaintiff Yturbide has also cross-appealed specifying error in the order of the court limiting costs for the travel expense of witnesses to expenses for travel for a distance of 100 miles from the place of travel. The action of the court was in accordance with the prevailing rule. It is stated, and the authorities therefor cited, in Kenyon v. Automatic Instrument Co., D.C., 10 F.R.D. 248 . See also Barnhart v. Jones, D.C., 9 F.R.D. 423 . There was no error in respect to the allowance of costs.

The judgments are affirmed.

LEMMON, Circuit Judge, did not participate in the decision of this case.

. It took form in hydrogen fluoride, silicone tetrafluoride, sodium aluminum fluoride, aluminum fluoride, and cryolite. The fluorine, as such is a gas which immediately combines with other elements to make these effluent compounds.

. From the Pre-Trial Orders: “Plaintiff contends that the poisonous fluoride compounds emanating from the plant of the defendant and carried by the prevailing winds upon and over, and settling upon, the lands upon which plaintiff resided and was present, were inhaled and were ingested by plaintiff, and said poisonous fluoride compounds rendered plaintiff permanently ill and caused plaintiff to suffer and to undergo serious and permanent personal injury.”

. During the trial in examining witnesses counsel for defendant made frequent reference to a British Government publication entitled “Medical Research Memorandum No. 22 on Industrial Fluorosis and State of Hazards to Man and Animal near Ft. William, Scotland.” This reports the result of a study of fluorosis affecting cattle and sheep near an aluminum factory. It reports with respect to such cattle grazing in the vicinity of the factory osteodystrophia, dental lesions, and “some cases of osteomalacia due to fluorine intoxication”; also severe loss of condition and loss of milk in affected animals.

. Dr. Donald Hunter, British physician and a director of research in industrial medicine of the Privy Council Medical Research Council, which produced the memorandum No. 22 referred to in footnote 3, supra, testified as follows: “Q. Why, Doctor, are both cattle and animals as well as men, humans, affected by the effluents from an aluminum factory? A. Because fluorides — fluorine compounds are deadly poisonous to mammalian tissues, and man is a mammal just as much as a cow or sheep.”

Error is assigned on account of the admission of evidence of damage to cattle. Since, as here indicated, that evidence had only a limited significance, it cannot be said that its admission was error. It was admissible for the purpose here stated.

. Dr. Hunter, a Fellow of the Royal College of Physicians, was director of a department for research of industrial medicine of the Medical Research Council, a public undertaking of the Privy Council. He had been called upon to give lectures in this field, at medical schools and before medical groups throughout the world. Before testifying he had examined and reviewed the reports and tests made in 1951, particularly the liver efficiency tests to determine the damage to the cells in the livers of the Martins made by Dr. Capps whom Dr. Hunter referred to as “probably the world’s greatest expert on this subject.”

. Dr. Capps described Paul Martin’s symptoms as follows: “In the first place, he developed diarrhea, indigestion, and heartburn, and bloating, particularly after meals. Those symptoms all were referable to the intestinal tract. And, secondly, he developed symptoms which we might refer to as skeletal symptoms: pain, particularly in his back, occasionally down the leg sufficiently severe so that he was unable to bend over and tie his shoes, for example, or even put his shoes on, ,as a matter of fact. These pains were aggravated by lifting and by various types of exertions. Tlie third group of symptoms had to do with the respiratory tract, the lungs. He developed a dry cough and shortness of breath particularly on exertion. And, finally he developed frequency of urination particularly at night, to such an extent that'— seven or eight times at night, and he was unable to sleep. Those were, in brief, the symptoms which he comjdamed of. And our observations in the hospital were directed towards trying to determine what those four groups were due to.”

After detailing at length the process by which he arrived at a diagnosis of what caused these symptoms which included a diagnosis of fluorosis, (a poisoning with fluorine), and toxic hepatitis, (toxic involvement of the liver), he stated his conclusion as follows: “In view, then, of this potential history of the exposure and in view of the fact that we were advised — that we were faced with a rather bizarre group of symptoms — here is an involvement of a number of the symptoms in the body which was unusual and definitely bizarre, and it corresponds exactly to the de *326 scription of published cases of fluorosis; furthermore, there is no other explanation. One cannot make another single diagnosis that would cover all of these symptoms. One could make four or five diagnoses, but, of course, that is always obviously a very poor thing to do. This is the picture of fluorosis. There is a history of the potential exposure. We are unable to find any explanation after a careful search, and under those circumstances one is justified, and not only justified, but you are forced to make the diagnosis of poisoning with fluorine.”

Dr. Hunter testified that “fluorine compounds are deadly poisons to mammalian tissues;” that when they enter the body they attack the enzymes in the body cells; that fluorine compounds fed in animal experiments produced degeneration of the liver and the kidney; that the symptoms of subacute fluorine poisoning were those which were seen in the Martins. It was his opinion that the Martin family were suffering from subacute fluorosis because their symptoms were identical with those in cases of that character which he had seen previously.

. This was an ironworks which produced the same fluoride effluence.

. See Martin v. Reynolds Metals Co., D. C., 135 F.Supp. 379 . The able trial judge, expressing doubt that the doctrine of Fletcher v. Rylands would be applied by the Oregon court to a case of this kind, held that the case should go to the jury upon the issue of negligence. Although in appropriate cases a judgment may be affirmed upon grounds other than those adopted in the court below, we think it questionable that such rule is applicable where the judgment is based upon the verdict of a jury acting upon instructions not comprehending the new grounds urged.

. The court cited and relied upon Suko v. Northwestern Ice & Cold Storage Co., 366 Or. 557 , 113 P.2d 209, 213, where the following language is used: “In the ordinary course of matters the mishap would not have occurred except through carelessness in the construction, inspection or use of the tank. And, as hereinabove stated, the premises on which the tank was located were in the exclusive possession and under the direct control of the defendant. Therefore, the doctrine of res ipsa loquitur applies.”

. The authors quote the following from Minutilla v. Providence Ice Cream Co., 50 R.I. 43, 49 , 144 A. 884, 887 , 63 A.L.R. 334 : “If the care commonly used by defendant had been exercised, the presence of these pieces of glass in the cream sold to plaintiff would have been impossible. Yet the fact remains that there they were. * * * ” (Footnote 6, p. 1106)

Prosser discusses the same matter (pp. 216-217) as follows: “But if the def end-ant merely offers evidence of his own acts and precautions amounting to reasonable care, it is seldom that a verdict can be directed in his favor. The inference from the circumstances remains in the case to contradict the evidence. If he testifies that he used proper care to insulate his wires, to inspect his chandelier, to drive his bus, or to keep defunct mice and wandering insect life out of his bottled beverage, the fact that electricity escaped from the wires, that the *331 chandelier fell, that the bus went into the ditch and the bug was in the bottle, with the background of common experience that such things do not usually happen if proper care is used, may permit reasonable men to find that his witnesses are not to be believed, that the precautions described were not sufficient to conform to the standard required or were not faithfully carried out, and that the whole truth has not been told. It is of course not impossible that proof of proper care may be so overwhelming as to call for a directed verdict, but in the ordinary case it will not be sufficient to destroy the inference from res ipsa loquitur.”

. Included in the instructions were the following:

. Sweeney v. Erving, 228 U.S. 233, 240 , 33 S.Ct. 416, 418 , 57 L.Ed. 815 : “In our opinion, res ipsa loquitur means that the facts of the occurrence warrant the inference of negligence, not that they compel such an inference; that they furnish circumstantial evidence of negligence where direct evidence of it may bo lacking, but it is evidence to be weighed, not necessarily to be accepted as sufficient; that they call for explanation or rebuttal, not necessarily that they require it; that they make a case to be decided by the jury, not that they forestall the verdict.”

. “May it please the Court, the defendant respectfully excepts and objects to the instructions given by the Court as to the applicability of the doctrine of res ipsa loquitur on the grounds that: one, that doctrine is not applicable here as a matter of law by reason of the fact that the claims asserted are phenomenal and unique in nature, and under the circumstances, the doctrine is not generally applicable in cases of that type; secondly, that this doctrine should not have been instructed, with respect to the res ipsa loquitur doctrine, because of the fact that the same is designed to give rise to an inference to take a case to the jury, and the evidence is clear here that there was no negligence, and the inference, if any, was rebutted by the affirmative proof with respect to the matter of negligence.”

. “Whether it would he reversible error for a trial court to instruct on the inference which arises in this type of case is a matter which need not now be decided.” Ritchie v. Thomas, supra, 224 P. 2d at page 551 .

. As Paula Martin was a minor there is no claim that the statute had run as to her.