dissenting:
I.
I thoroughly disagree with the majority’s conclusion that the district court erred in excluding Dr. Jenkins’ opinion that Mr. Moore’s reactive airway disease (“RAD”) was triggered by his exposure to a Toluene solution at Ashland’s facility.
The question we must decide is whether the district court was overzealous in performing the “gatekeeper” role the Supreme Court assigned to it in Daubert1 to admit only expert opinions that are “reliable.” Daubert explains that expert testimony must be “scientific,” that is based on scientific “knowledge” that is “grounded in the methods and procedures of science,”2 and that assists the trier of fact by having a “valid scientific connection to the pertinent inquiry.” Id. at 590-92, 113 S.Ct. at 2795-2796; see also G. Michael Fenner, The Daubert Handbook: The Case, its Essential Dilemma, and its Progeny, 29 Creighton L.R. 939 (1996). At bottom, the district court was charged with making an assessment of whether the reasoning and methodology used by Dr. Jenkins was scientifically valid and whether that reasoning properly applied to the facts at hand.
As in other evidentiary questions, the proponent of the expert testimony must satisfy the trial judge by a preponderance of the evidence that the Daubert conditions have been met. Claar v. Burlington Northern R. Co., 29 F.3d 499 (9th Cir.1994). We review preliminary factual findings of the district court necessary for determining admissibility for clear error. See Bourjaily v. United States, 483 U.S. 171, 175, 107 S.Ct. 2775, 2778, 97 L.Ed.2d 144 (1987). We reverse a district court’s evidentiary ruling only for manifest abuse of discretion. Allen v. Pennsylvania Eng’g. Corp., 102 F.3d 194 (5th Cir.1996).
The expert opinion at issue here is, of course, the testimony of Dr. Jenkins on the causal connection between Mr. Moore’s exposure to chemicals and his asthmatic-type condition, RAD. Dr. Jenkins is a well-qualified physician specializing in internal medicine and pulmonary disease. Mr. Moore was referred to Dr. Jenkins by his attorney for examination and evaluation. Dr. Jenkins saw Mr. Moore on three occasions, took a history from him, examined him, and gave him a number of tests.
The court did permit Dr. Jenkins to testify concerning his contact with Mr. Moore and to give a diagnosis and prognosis for Mr. Moore’s recovery. However, the district court declined to permit Dr. Jenkins to testify concerning the cause of Mr. Moore’s condition. Before declining to allow Dr. Jenkins to give an opinion on the causal connection between the chemical exposure and Mr. Moore’s condition, the district court allowed Mr. Moore to proffer Dr. Jenkins’ live testimony. The district court also considered Dr. Jenkins’ deposition, which had been submitted earlier in support of an in limine motion to exclude his testimony.
My review of the record leads me to conclude that the district court was fully justified in excluding the testimony. The district court’s reasons for excluding part of Dr. Jenkins’ testimony were adequately spelled out in the record. R. at 10.154-56. The district court found that Dr. Jenkins had no information concerning the level or duration of Mr. Moore’s exposure to the chemicals. This finding is fully supported by the record. Dr. Jenkins himself admitted that he did not know the size of the trailer where the spill occurred or the extent of the spill. The district court was entitled to conclude from this that Dr. Jenkins’ estimate that the air in the trailer contained over 200 parts per million of the spilled chemicals was nothing more than speculation. Furthermore, at his deposition and at the pretrial motion in limine, Dr. Jenkins could not point to one piece of scientific literature or research linking exposure to the spilled chemicals and *711RAD. The majority’s conclusion rests on a number of premises that will not withstand scrutiny, • each of which I consider below.
II.
A.
The majority begins with the remarkable premise that clinical medicine is not “hard” science; ergo, a physician’s opinion that an illness was caused by a patient’s exposure to a toxic substance is not a “scientific” expert opinion. This leads the majority to conclude that Daubert’s teaching is inapplicable to the issue presented to the trial court: Whether to admit Dr. Jenkins’ testimony. We need go no further than Daubert itself to demonstrate the fallacy of this conclusion. In Daubert, the Supreme Court considered whether the district court erred in rejecting testimony proffered by the plaintiff to establish a causal connection between the plaintiffs exposure to the drug Bendictin and birth defects. The subject matter of these witnesses’ testimony, medical causation, was obviously considered “scientific” by the Court: It was the factual predicate to the Court’s guidelines for evaluating whether to admit expert testimony. Although the tendered witnesses in- Daubert were not physicians, the focus of the decision is on the subject matter of the testimony. The Court does not suggest that its guidelines would not apply if the bearer of the opinion on medical causation had been a physician rather than a biologist or chemist.
If reinforcement is needed for the proposition that medical causation testimony of physicians is indeed “scientific” testimony, it can be found in our post-Daubert decisions and all of the decisions addressing the issue from sister circuits.
In Allen v. Pennsylvania Eng’g. Corp., 102 F.3d 194 (5th Cir.1996), this Court held that expert testimony regarding exposure to chemicals and medical causation of cancer was not scientifically valid. The proffered experts’ testimony consisted of human epidemiological evidence suggesting a link between ethylene oxide (EtO) exposure and increased risk of brain cancer, scientific studies conducted on rats, and the fact that EtO is known as a mutagen and genotoxin. Id. at 196. There were absolutely no scientific studies on a link between human brain cancer and EtO exposure. Id. at 197. In excluding the proffered expert testimony, the court stated emphatically that
the goal of Daubert and this court’s previous eases has been to bring more rigorous scientific study into the expression of legal opinions offered in court by scientific and medical professionals. In the absence of scientifically valid reasoning, methodology and evidence - supporting these experts’ opinions, the district court properly excluded them.
Id. at 198 (emphasis added).
In Wheat v. Pfizer, Inc., 31 F.3d 340 (5th Cir.1994), the plaintiff sought to offer the testimony of a doctor to support his claim that the drug Feldene caused the plaintiff’s hepatitis. While the admissibility of the expert’s testimony was rendered moot by the court’s resolution of the case on other grounds, the court noted “in passing that [the doctor’s] testimony would not have survived the test of Daubert v. Merrell Dow Pharmaceuticals, Inc.” Id. at 343 (citation omitted). The court continued:
At the hearing held to evaluate his proffered testimony, Dr. George hypothesized that the combination of Feldene and Chlorzoxazone may have caused [the plaintiff’s] hepatitis. He admitted, however, that no study of the combined effects of the drugs had ever been done, and thus his hypothesis lacked an empirical foundation. Neither had it been subjected to peer review and publication, which Daubert also identifies as key.
Id. at 343 (citation omitted).
Finally, in Carroll v. Morgan, 17 F.3d 787 (5th Cir.1994), a case cited by the majority, we considered whether the district court abused its discretion in allowing a cardiologist to give his opinion on the cause of the plaintiffs death. The court recognized, that Daubert controlled the analysis. Id. at 789-90.
Our sister circuits are in agreement that medical causation testimony by physicians is indeed “scientific” expert testimony. Holbrook v. Lykes Bros. S.S. Co., Inc., 80 F.3d *712777 (3d Cir.1996) (testimony of plaintiffs treating physician on diagnosis and causes of plaintiffs cancer subject to Daubert as scientific evidence); Cavallo v. Star Enter., 100 F.3d 1150 (4th Cir.1996) (doctors’ opinions on cause of plaintiffs’ diseases properly excluded under Daubert as not being scientifically reliable); Glaser v. Thompson Med. Co., Inc., 32 F.3d 969 (6th Cir.1994) (physician’s testimony as to cause of plaintiffs injuries properly admitted because of valid scientific basis under Daubert); Hose v. Chicago Northwestern Transp. Co., 70 F.3d 968 (8th Cir.1995) (physician’s expert opinion on plaintiffs exposure to toxic fumes and dust as causing disease subject to Daubert factors); Joiner v. General Elec. Co., 78 F.3d 524 (11th Cir.1996), cert. granted, -U.S.-, 117 S.Ct. 1243, 137 L.Ed.2d 325 (1997) (plaintiffs experts’ opinions on cause of lung cancer deemed “scientific knowledge”); Raynor v. Merrell Pharm. Inc., 104 F.3d 1371 (D.C.Cir.1997) (doctor’s testimony regarding cause of birth defects governed by Daubert factors); Cella v. United States, 998 F.2d 418 (7th Cir.1993) (case decided before Daubert but recognizing that expert medical opinion is “scientific” and should have “an epidemiological or scientific foundation”).
In sum, all our post-Daubert cases, along with those of our sister circuits, consistently recognize that the admission of a physician’s testimony on medical causation is governed by Daubert’s requirements, thus announcing in a voice that is loud and clear that such testimony is indeed “scientific” expert testimony. The majority has not cited a single federal appellate case to support its contention that a physician’s testimony on medical causation is not considered “scientific” expert testimony.
If somehow one accepts the majority’s view that Dr. Jenkins’ testimony was not “hard” “scientific” expert testimony, it is nonetheless controlled by Daubert. In Watkins v. Telsmith, 121 F.3d 984 (5th Cir.1997), this Court concluded that the Daubert analysis applied to proffered expert testimony of an engineer, based on his training and experience, regarding the design of a conveyor. We expressly rejected the holding of the majority in today’s -case and the position of the Tenth Circuit that “application of the Daubert factors is unwarranted in cases where expert testimony is based solely on experience or training.” Id. at 989 (quoting Compton v. Subaru, 82 F.3d 1513 (10th Cir.1996)).
B.
The majority next concludes that we owe no deference to the district court’s evidentiary ruling. This determination is based on the conclusion that the district court clearly erred or was confused in its determination of precisely what chemicals were spilled from the drum and inhaled by Mr. Moore. The record does not justify this conclusion.
The Material Safety Data Sheet (“MSDS”) was introduced by the plaintiff and was the central item of documentary evidence in the case. This document listed each chemical, including Toluene, that made up the contents of the drum. The MSDS gave special warning about exposure to Toluene, and early in the litigation Mr. Moore focused on Toluene. This is clear from the pleadings of both parties as well as the pretrial depositions of Drs. Jenkins and Alvarez, both of whom had been furnished with copies of the MSDS. Because the MSDS warned specifically about exposure to Toluene and the physicians focused on Toluene, counsel then concentrated on Toluene, the chemical the physicians thought was important. Following the deposition and just before trial, the physicians submitted affidavits designed to expand their testimony to say that they relied on the entire mix of chemicals in the drum rather than Toluene alone as the causative agent.
The majority’s numerous references to the trial court as confused as to whether Moore had been exposed to a single chemical or to a mixture of chemicals; lacking a full understanding of both Dr. Jenkins’ testimony and the chemical contents of the leaking drum; and “laboring] under confusion” are not supported by a fair review of the record. Excerpts from the record, reproduced in the footnote below, demonstrate that the trial judge was fully aware of the chemicals to which Mr. Moore was exposed at Ashland’s *713facility.3 The record simply does not-justify the majority’s refusal to give the deference due the trial court in excluding the evidence.
C.
As an independent basis for excluding the evidence, the district court was entitled to conclude that the expert’s opinion was not relevant to the facts at hand. The record does not establish either the level of the chemicals that Mr. Moore breathed or the level required to cause RAD.
In Allen, 102 F.3d at 194, the plaintiff offered opinion testimony that his brain cancer had been triggered by his employment-related contact with cylinders containing ethylene oxide. There was a paucity of evidence relating to the extent and level of Allen’s exposure to the alleged harmful chemical. As one ground for affirming the district court’s rejection of this causation testimony, the court stated: “Scientific knowl-. edge of the harmful level of exposure to a chemical, plus knowledge that plaintiff was exposed to such quantities, are minimal facts necessary to sustain the plaintiffs burden in a toxic tort case.” Id. at 199.
The Allen court cited with approval Wright v. Willamette Industries, 91 F.3d 1105 (8th Cir.1996). In that case, the plaintiffs lived a short distance from the defendant’s fibreboard manufacturing plant. Plaintiffs sought damages for injuries they argued were caused by breathing airborne formaldehyde and other harmful chemicals emitted from the plant. The court reversed the district *714court’s ruling allowing expert opinion testimony that the plaintiffs’ complaints were related to their exposure to the plant’s emissions. The court explained that in a suit to recover damages in a tort action:
It is therefore not enough for a plaintiff to show that a certain chemical agent sometimes causes the kind of harm that he or she is complaining of. At a minimum, we think that there must be evidence from which the factfinder can conclude that the plaintiff was exposed to levels of that agent that are known to cause the kind of harm that the plaintiff claims to have suffered. See Abuan v. General Elec. Co., 3 F.3d at 333.
Id. at 1107.
D.
Even if the district court had found that Dr. Jenkins had sufficient evidence of the level of chemicals to which Mr. Moore was exposed, the district court was entitled to conclude that Dr. Jenkins had no scientific, technical, or other specialized knowledge that would assist the trier of fact in resolving the causation question. Dr. Jenkins’ conclusion that the Toluene solution triggered Mr. Moore’s RAD was based on two assumptions: 1) any chemical with irritating properties can cause RAD (R. at 10.132); 2) the manufacturer’s general warning contained in the MSDS that exposure to the chemical could cause damage to a number of organs, including the lungs, established that Toluene was an irritant. The temporal connection between Mr. Moore’s exposure to Toluene and his onset of symptoms clinched Dr. Jenkins’ conclusion.
Dr. Jenkins could cite no scientific support for his conclusion that exposure to any irritant at unknown levels can trigger this asthmatic-type condition. Surely a court is entitled to view such an unsupported, unscientific generality with skepticism. The purpose of Daubert was to exclude such speculation, based primarily on a temporal connection, as lacking any scientific validity.
Dr. Jenkins admitted that he knew nothing about who prepared the MSDS, what tests were conducted to support them, or the warning label on the drum of Toluene. No scientific foundation was laid to support the notion that this general warning could serve as the basis for concluding that exposure to unknown quantities of Toluene would likely cause RAD.4 Dr. Jenkins had no personal experience with Toluene. He admitted that Mr. Moore was the first patient he had examined who claimed that he suffered injury from exposure to this chemical.
The scientific proof on causation in this ease suffers the same infirmities as the causation evidence in Wright, in which the court stated:
It is true that Dr. Frank Peretti, after a great deal of prodding, testified that the Wrights’ complaints were more probably than not related to exposure to formaldehyde, but that opinion was not based on any knowledge about what amounts of wood fibers impregnated with formaldehyde involve an appreciable risk of harm to human beings who breathe them. The trial court should therefore have excluded Dr. Peretti’s testimony as Williamette requested it to do, because it was not based on scientific knowledge, [citing cases] Dr. Peretti’s testimony regarding the probable cause of the Wrights’ claimed injuries was simply speculation.
The jury could therefore only have speculated about whether the amount of formaldehyde from Williamette’s plant to which each plaintiff was exposed was sufficient to cause their injuries or, indeed, any injuries at all____ Without proving hazardous levels of exposure to Williamette’s *715formaldehyde, the Wrights failed to carry their burden of proof at trial on the issue of causation because the evidence failed to support a reasonable inference in favor of the jury’s implicit finding against Williamette on the causation issue.
Wright, 91 F.3d at 1108.
The district court also was entitled to consider uncontroverted evidence of other possible causes of Mr. Moore’s asthmatic condition. First, Mr. Moore had just returned to work after recovering from pneumonia. Second, Mr. Moore was a cigarette smoker. He had smoked about a pack of cigarettes a day for twenty years. Finally, Mr. Moore admitted that he had suffered from asthma as a child.
E.
I agree with the majority that Dr. Jenkins is a well-qualified pulmonary specialist. But I disagree with the majority’s suggestion that when we consider whether an expert has given a scientific opinion we have a sliding scale where the greater the witness’ qualifications the less scientific the evidence must be to support the opinion.
In Rosen v. Ciba-Geigy Corp., 78 F:3d 316 (7th Cir.1996), the Seventh Circuit considered the admissibility of a highly qualified medical expert’s opinion on causation that lacked formal scientific support. In that case, the plaintiff sought to connect his use of a nicotine patch, to help him stop smoking, to his sudden heart attack. A distinguished cardiologist and department head at the University of Chicago testified that the heart attack was indeed triggered by the use of the nicotine patch. The Seventh Circuit, speaking through Judge Posner, held that the district court correctly declined to permit this testimony because it had an inadequate scientific basis. The court stated: “Under the regime of Daubert a district judge asked to admit scientific evidence must determine whether the evidence is genuinely scientific, as distinct from being unscientific speculation offered by a genuine scientist.” Id. at 318(ci-tation omitted).
In summary, I am satisfied that the district court correctly rejected — and certainly did not abuse its discretion in excluding — Dr. Jenkins’ testimony as being without sufficient factual or scientific foundation.
F.
Finally, I am satisfied that even if the district court erred in rejecting this evidence, it was harmless error. As outlined in the majority opinion, the district court permitted Dr. Jenkins to testify about his examination and testing of Mr. Moore. The district court also admitted Dr. Jenkins’ conclusion that Mr. Moore was suffering from RAD, along with his prediction for future treatment and disability.
Dr. Jenkins’ and Dr. Alvarez’s testimony was completely consistent except that Dr. Jenkins’ examination and tests were arguably more comprehensive than Dr. Alvarez’s. The district court allowed the plaintiff to produce evidence of Dr. Jenkins’ examination and tests, and Dr. Alvarez accepted Dr. Jenkins’ findings as accurate. On the causation issue, Dr. Jenkins’ proffered testimony was virtually identical to Dr. Alvarez’s testimony. Both testified that the Toluene solution was an irritant, that the MSDS established this fact, and that the temporal connection between Mr. Moore’s exposure to Toluene and his onset of RAD justified the conclusion that the two were related. Dr. Jenkins suggested no material factual or scientific basis for his opinion on causation that Dr. Alvarez did not rely on. Dr. Jenkins testified that he had never had a previous patient who claimed that Toluene triggered his RAD and he admitted that he conducted no tests on this question. I am unable to see why a jury would have gleaned any assistance from Dr. Jenkins’ testimony on the critical causation issue that it did not find in Dr. Alvarez’s testimony. A district court’s refusal to permit a party to call two expert witnesses to give cumulative testimony is rarely considered prejudicial error.
As one court observed, where the excluded expert’s testimony does not add a new angle or argument to the point at issue, the testimony is considered cumulative and its exclusion is harmless error. Kendra Oil & Gas, Inc. v. Homco, Ltd., 879 F.2d 240 (7th Cir.1989); see also Collins v. Wayne Corp., 621 *716F.2d 777, 782 (5th Cir.1980); Miley v. Delta Marine Drilling Co., 473 F.2d 856, 858 (5th Cir.1973). Mr. Moore’s position on causation was presented by Dr. Alvarez; he was not entitled to have that same position repeated by Dr. Jenkins.
Nov. 12, 1997III.
The Supreme Court has directed the district courts to control with a firm hand expert testimony to prevent litigation abuse so familiar to all of us. The district court took a careful look at Dr. Jenkins’ testimony, applied the correct standard, and excluded the testimony. After reviewing the record, I fail to see how the district court could have reached any other conclusion. The majority’s “let it all in” view sends exactly the wrong message to conscientious district courts. I therefore dissent.
Before KING, JOLLY, HIGGINBOTHAM, DAVIS, JONES, SMITH, DUHÉ, WIENER, BARKSDALE, EMILIO M. GARZA, DeMOSS, BENAVIDES, STEWART, PARKER and DENNIS, Circuit Judges.*ORDER
BY THE COURT:A majority of the judges in active service having determined, on the court’s own motion, to rehear this ease en banc,
IT IS ORDERED that this cause shall be reheard by the court en banc with oral argument on a date hereafter to be fixed. The Clerk will specify a briefing schedule for the filing of supplemental briefs.
. 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993).
. 509 U.S. at 589-90, 113 S.Ct. at 2794-2795.
. During the final pretrial conference, held three days before trial, the following exchange occurred:
THE COURT: Isn’t that the only chemical at issue though?
MR. BLACK: No. No, Your Honor,
THE COURT: What chemical—
(Talking off the record)
MR. BANOWSKY: The product was naphtha, that I could recall — naphtha, some propylene glycol methyl ethers, toluene, and naphtha.
THE COURT: But your loss is premised on — is it toluene, what is it?
MR. BLACK: Toluene.
THE COURT: Toluene. The loss is premised on toluene, isn’t it?
MR. DAVIS: Your Honor, it’s generically referred to as toluene throughout this litigation; however, what this chemical is and what everybody knows in this case from the MSDS is that it is Dow Coming 125-35 release coated which is a component mixture of various chemicals. It is a chemical—
THE COURT: Okay. Let me stop you.
MR. DAVIS: — release coating used.
THE COURT: You’re saying that whatever chemical is listed in the Dow Chemical MSDS as toluene is what’s at issue here.
MR. DAVIS: No. All the chemicals are at issue here because it's-the entire — this release coating is what spilled in the truck. The release coaling is made up of toluene, naphtha, petroleum dystolate [sic] and propylene glycol methylene.
THE COURT: All right. That's what I’m trying to determine. There is evidence in the record that these other chemicals were involved in the release? That's all I want to know.
MR. DAVIS: Yes.
THE COURT: Okay. All right. Go on, Mr. Black.
R. at 6.23-24.
The following two excerpts are of testimony given at a hearing outside of the jury’s presence on the motion in limine to exclude Dr. Jenkins’ testimony:
BY MR. GREEN:
Q All right. After you looked at the MSDS sheet that was based upon testing, took the history, you performed the test, you did the examination. Did you come to a conclusion as to the cause of Mr. Moore's reactive airways disease?
A [by Dr. Jenkins] Yes.
Q And what was that conclusion?
A Well, I feel it Was the chemical substances to which he was exposed. There are several' of them. ..
Q All right. And what was that?
A Well, toluene, naphtha, propylene glycol methyl ether I think were the principal ones that had irritating properties.
R. at 10.132.
THE COURT: ... Mr. Carpenter, while Dr. Jenkins is looking at that, Mr. Green, would you listen to this and tell me if this is the correct rendition of the chemicals you asked Dr. Jenkins about? Naphtha — And what else?
MR. GREEN: Naphtha, toluene, ethyl methyl glycol ether, I believe.
R. at 10.135-36.
We can tell frpm the trial judge’s reasons for excluding Dr..Jenkins testimony that she had not forgotten the MSDS or /'the substance” it described: ' ■
Last but not least, I don't know why, if his entire causation testimony is based upon the MSDS, that his testimony would be necessary since the MSDS is in evidence and the MSDS recites that under certain circumstances this substance may be-.irritant to throat and nose, and vapors may injure, quote, -“blood, liver, lungs, kidney, and nervous system depending upon the degree and the effects of concentration and length of exposure.”
R. at 10.155.
. The only literature Dr. Jenkins arguably relied on was an article by Brooks on several case studies. One of these studies related to a 19-year-old store clerk’s exposure to floor sealant containing, among other things, Toluene. In addition to the scientific limitations of a single case study, the circumstances surrounding that study were markedly different from Mr. Moore’s exposure to the chemical. The floor sealant to which the store clerk was exposed contained a number of irritants in addition to Toluene. Some of the other ingredients may have been more irritating than Toluene. Also, the clerk was exposed to Toluene while working in a small space for two and one-half hours. While he was still on the job, the clerk experienced a number of symptoms, such as dizziness and headaches.