with whom PARKER and STEWART, Circuit Judges, join, dissenting:
I respectfully dissent.
The majority en banc opinion (1) conflicts with the view of other circuits, a state court of last resort, and scholarly commentary, in *280holding that (a) a clinical medical expert cannot express an opinion as to a causal relationship between a chemical compound and a plaintiffs disease, although the opinion is based on the sound application of generally accepted clinical medical methodology, unless the causal link is confirmed by hard scientific methodology as per the Daubert factors1, see Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 593-94, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993); (b) the temporal relationship between chemical exposure and symptoms of disease are to be accorded little weight by courts in assessing an expert’s determination of causation with either clinical medical or hard science methodology; (c) even when an expert has hard scientific support for a general causal relationship between a chemical compound and a particular disease, his opinion of a specific causal relationship between the compound and an individual’s disease is “suspect” unless the expert also has scientifically accurate data as to the level of that person’s exposure to the chemical compound; (2) conflicts with Supreme Court decisions by conducting a de novo trial of the preliminary assessment hearing on the record, substituting its own erroneous ruling and reasons for those of the district court, and disregards the district court’s errors of law, clearly erroneous factual findings, and abuse of discretion.
1.
After Daubert, federal courts have become balkanized on important questions that confront federal trial judges daily, e.g., whether Daubert applies outside the field of hard science; if so, whether Daubert’s gatekeep-ing function applies to the admission of any or all of the other types of expert testimony; if so, whether application of the Daubert “factors” is required in the admission of any or all testimony based on knowledge not derived by hard scientific methodology. Even before the present en banc circuit opinion there was a clear and present need for the Supreme Court to clarify whether and, if so, how, Daubert applies to expert testimony based on knowledge derived by disciplines or sources other than the hard sciences. E.g., 29 Charles A. Wright and Victor J. Gold, FEDERAL PRACTICE AND PROCEDURE § 6266 (1997); 2 Michael H. Graham, HANDBOOK of Federal Evidence § 702.5, pp. 22-26 (Supp. 1998).
(a)
The majority opinion represents an eccentric additional fragmentation of the Daubert picture that underscores the need for Supreme Court guidance. This circuit now takes the position that a clinical medical expert, correctly using and applying generally accepted clinical medical methodology, may not express an opinion as to whether a particular chemical compound caused, aggravated, or contributed to a person’s disease or disorder unless that opinion is corroborated by hard scientific methodology that passes muster under a rigid application of the Dau-bert factors.
The majority’s rule applies even to single plaintiff negligence actions that do not involve substances alleged to cause diseases in large numbers of persons or diseases having long latency periods. The en banc majority opinion emanates from a case in which a single plaintiff claims to have'developed a reactive airways disorder as a result of a defendant’s negligence in causing him to clean up a spillage of a chemical compound without taking any safety precautions. The defendant refused to provide the plaintiff with a respirator or to measure the air contamination with a safety meter although the defendant had both devices ready at hand. The plaintiff was required to work in and around an enclosed 28-foot trailer for about an hour in cleaning up the spilled chemical compound.
*281Unlike many toxic torts situations, in Mr. Moore’s case there was not a long latency period between the onset of symptoms and the chemical compound gases that were alleged to have caused his illness. The onset of the plaintiffs respiratory disease occurred less than an hour after his exposure during his clean up of the chemical compound. He immediately sought emergency medical treatment, which included being given oxygen, and he has been under treatment for his respiratory disease ever since. The particular circumstances of the plaintiffs inhalation injury, combined with the fact that so few humans have ever been subjected to a similar exposure to the chemical compound involved, obviously impacted on the manner in which the plaintiff could prove causation. The quantity of persons who sustain this type of exposure was simply too small for a plaintiff to be able to provide epidemiological, animal testing or other hard scientific evidence linking the particular chemical compound to reactive airways disease. See Zuchowicz v. United States, 140 F.3d 381, 385-86 (2nd Cir.1998)(described infra.).
Although the en banc majority recognizes that cases involving chemical compounds which have not been subjected to hard scientific testing must be timely resolved and cannot await the fortuity of relevant scientific experimentation, the majority nevertheless insists that every admissible medical causation opinion in a chemical injury case must have a hard science, Daubert factor related basis. If such hard scientific data is not available, the majority decrees, a plaintiff must face trial or the defendant’s summary judgment motion without a medical causation expert witness.2
*282The majority opinion creates a schism between this court and other circuits and a state court of last resort and disregards the teachings of federal evidence law scholars.
The Second, Fourth, and Third Circuits have held that a clinical physician may, consistently with Daubert, express an opinion, based on clinical medical methodology generally accepted within that discipline, that a particular toxic substance caused the patient’s disease or death, without hard scientific corroboration under an inflexible application of the Daubert factors.
The Second Circuit in McCulloch v. H.B. Fuller Co., 61 F.3d 1038 (2nd Cir.1995), rejected the defendant’s argument for exclusion of a clinical physician’s opinion, as scientifically unfounded, that glue fumes caused the plaintiffs respiratory symptoms and throat polyps. The doctor’s opinion was based entirely upon his use of clinical medical methodology, without any hard science or strict Daubert factor related basis. The doctor could not point to a single piece of medical literature that said that glue fumes cause throat polyps. In describing the doctor’s use of clinical medical methodology as vouching for the reliability of his opinion, the court stated:
[Dr.] Fagelson based his opinion on a range of factors, including his care and treatment of McCullock; her medical history (as she related it to him and as derived from a review of her medical and surgical reports); pathological studies; review of Fuller’s MSDS; his training and experience; use of a scientific analysis known as differential etiology (which requires listing possible causes, then eliminating all causes but one); and reference to various scientific and medical treatises. Disputes as to the strength of his credentials, faults in his use of differential etiology as a methodology, or lack of textual authority for his opinion, go to the weight, not the admissibility, of his testimony. Id. at 1044.
In Zuchowicz v. United States, 140 F.3d 381 (2nd Cir.1998), the Second Circuit reaffirmed its holding in McCulloch The Zu-chowicz court approved the admission of a pulmonary medical expert’s opinion that a negligent overdose of Danocrine had been responsible for the pulmonary disease related death of the plaintiffs wife. The doctor based his opinion on the temporal relationship between the overdose and the start of the disease, the deceased’s apparent good health prior to the overdose, and the differential etiology method of excluding other possible causes. Id. at 385. He also testified that Mrs. Zuchowicz’s illness was similar in onset, timing and course of development to other cases of pulmonary diseases known to have been caused by other classes of drugs. Id. at 385-86. There had been no scientific tests to determine the effects of dosages at the level received by Mrs. Zuchowicz, and the doctor’s opinion as to medical causation, based solely on clinical medical methodology, was not confirmed by any hard science or strict Daubert factor evidence. See also Am-brosini v. Labarraque, 101 F.3d 129, 138 (D.C.Cir.l996)(stating that the fact that a case may be the first of its type should not prevent a plaintiffs doctor from testifying as to causation).
Similarly, the Fourth Circuit in Benedi v. McNeil-P.P.C., Inc., 66 F.3d 1378, 1384 (4th Cir.1995), upheld the plaintiffs recovery for severe liver damage resulting from his use of Extra-Strength Tylenol contemporaneously with alcohol due to the manufacturer’s negligent failure to warn. The Court of Appeals rejected McNeil’s argument that the medical causation testimony of the plaintiffs clinical physicians based on the methodology of their discipline, such as the microscopic appearance of his liver, the Tylenol found in his blood, the history of several days of using Tylenol and alcohol, the liver enzyme blood level, and the lack of evidence of a viral or other cause of liver failure, was unreliable because they did not have or rely on epidemiological data. The Benedi court stated: “We will not declare [the clinical medicine] methodologies invalid and unreliable in light of the medical community’s daily use of the same methodologies in diagnosing patients.” Id.; see also, Maryland Casualty Co. v. Therm-O-Disc, Inc., 137 F.3d 780, 785 (4th Cir.1998)(“[T]his circuit has taken the position that the Daubert court ‘was not formulating a rigid test or checklist,’ and was ‘relying *283instead on the ability of federal judges to properly determine admissibility.’ ”)(citing and quoting Benedi 66 F.3d at 1384).
The Third Circuit in In Re Paoli R.R. Yard PCB Litigation, 35 F.3d 717 (3rd Cir.1994) held that a clinical physician’s methodology of differential diagnosis was sufficiently reliable to support the admissibility of that expert’s opinion that polychlorinated biphe-nyls (PCBs) caused specific plaintiffs’ illnesses. The Paoli court, heeding Daubert’s admonition that the inquiry as to whether a particular technique or method is reliable is a flexible one, id. at 742, reasoned that “differential diagnosis can be considered to involve the testing of a falsifiable hypothesis (e.g. that PCBs caused a plaintiffs cancer) through an attempt to rule out alternative causes,” and although it “involves assessing causation with respect to a particular individual[,][t]his merely makes it a different type of science than science designed to produce general theories; it does not make it unreliable science.” Id. at 758. Moreover, the Pao-li court concluded that a clinical physician’s performance of standard diagnostic techniques provides prima facie evidence that a doctor has considered alternative causes and has attempted to test his or her initial hypothesis as to cause. Id.
The Court of Criminal Appeals of Texas, a state court of last resort, in Nenno v. State, 970 S.W.2d 549, 1998 WL 331283 (Tex.Crim.App. June 24, 1998)(“This opinion has not been released for publication in the permanent law reports. Until released, it is subject to revision or withdrawal.”), in reviewing the defendant’s capital murder conviction and death sentence, held that the trial court did not err in finding reliable and admitting the state’s future dangerousness expert’s opinion that the defendant would be a threat to society. The expert, an FBI agent who specialized in studying the sexual victimization of children, based his opinion on his study of over 1,000 cases, personal interviews with inmates convicted of child sex offenses, examination of inmates’ psychological records, and study of the facts of the offenses involved. The Nenno court rejected the defendant’s argument that the expert’s opinion was not reliable because it did not rely on criteria substantially identical to the Daubert factors. Instead, the Nenno court concluded that “the four factors listed in Daubert do not necessarily apply outside of the hard science context; instead methods of proving-reliability will vary, depending upon the field of expertise.” Id. at 561 (citing the panel opinion in the present case, Moore v. Ashland Chemical, Inc., 126 F.3d 679, 685-689 (5th Cir.1997)).
Although the Nenno decision did not involve the testimony of a clinical physician as to cause of disease in a specific person, the court relied directly upon the Moore panel decision and its underlying principle that the reliability of an expert witness’s opinion ordinarily should be judged by whether it is soundly grounded in the methodology of the expert’s discipline. Thus, Nenno, which permits experts to predict the future causation of criminal harm by a specific person without the support of any hard scientific, strict Dau-bert factor type methodology, is at odds with the premise of the present en banc majority opinion.
In similar manner, additional federal circuit decisions conflict in principle with the en banc majority opinion’s insistence on an inflexible, unthinking application of the Dau-bert factors to expert opinions based on knowledge and methodology outside the realm of hard science. E.g., Tyus v. Urban Search Management, 102 F.3d 256, 263 (7th Cir.1996)(“Social science testimony, like other expert testimony ... must be tested to be sure that the person possesses genuine expertise in a field and that her testimony adheres to the same standards of intellectual rigor that are demanded in her professional work.”) (internal quotation marks and brackets omitted); Hose v. Chicago Northwestern Transp. Co., 70 F.3d 968, 974 (8th Cir.1995)(clinical physician’s opinion that patient’s inhalation of manganese caused patient’s manganese encephalopathy was reliable although based only on patient history, laboratory studies of manganese levels in patient’s body and work clothes, clinical examinations, a series of MRIs, and other doctors’ reports); United States v. Jones, 107 F.3d 1147 (6th Cir.1997)(although Daubert’s gatekeeper function is applicable to all expert *284testimony, the Daubert factors do not extend outside the hard scientific orbit to handwriting experts); see also Tassin v. Sears, Roebuck and Co., 946 F.Supp. 1241, 1247-48 (M.D.La.1996)(holding that for an expert’s opinion to be considered reliable he must use the methodology of experts in his particular field).
The majority’s opinion requiring a rigid, mechanical application of the Daubert factors beyond the ambit of the hard sciences also conflicts with the views of leading scholars, jurists and practitioners.3 For example, the report of the American College of Trial Lawyers on Standards and Procedures For Determining the Admissibility of Expert Evidence After Daubert, 157 F.R.D. 571 (1994) recognizes that the basic Daubert requirement that a trial judge determine whether a proffer of expert testimony is reliable or valid applies to all forms of expert testimony and that the particular expert at issue should have her methodology, i.e. the validity of her opinion, judged by the principles applicable to “that particular field.” Id. at 577. In regard to the specific Daubert factors which the majority so rigidly applies, the American College of Trial Lawyers’ report concludes that:
... Justice Blackmun’s “general observations” about the factors that a federal judge ought to consider in evaluating the soundness of scientific methodology, set forth in part II-C of his opinion, are specifically aimed at the evaluation of scientific testimony. Of course, some of these factors may be highly relevant to an evaluation of certain types of non-scientific expert evidence. For example, whether the proffered methodology can be and has been tested may very well be pertinent to an examination of non-scientific but “technical” expert evidence. Peer review and publication may be an important factor with respect to testimony involving social sciences. And the “general acceptance” of a methodology within a particular discipline will be crucial in many cases. The point is that any one of Justice Black-man’s four factors may or may not have applicability to proffers of non-scientific expert evidence. The inquiry to be made concerns the fundamental principles by which the validity of a methodology is to be judged in the particular field of knowledge. Id. (footnotes omitted)(emphasis added)
Leading federal evidence commentators have noted that the Daubert opinion is ambiguous and has given rise to a number of interpretations. E.g. 29 Charles A. Wright and Victor J. Gold, Federal PRACTICE and ProCEdure § 6266 (1997). They observe that at its narrowest Daubert can be read to *285allow judges to exercise a significant gatek-eeping function only in the case of expert testimony in the hard sciences based on novel theories and methodologies. Id. at 289. They further state that the broadest reading of Daubert is that it applies to all reliability issues presented by all expert testimony. Id. at 290. In rejecting the broadest view, Wright and Gold state:
This broadest interpretation of Daubert should be rejected. As noted above, it is inconsistent with both policy and precedent to make the admissibility of all expert testimony depend upon a showing that the expert’s testimony is completely reliable in every respect. Since Daubert does not explicitly take such a position, and nothing in the Evidence Rules compels it, it seems unlikely that the Court intended such a departure from past practice. In overturning Frye, it is unlikely that the Court in Daubert sought to make the admission of scientific evidence harder. Id. at 290-91 (footnotes omitted).
Professor Michael Graham contends that Daubert boxes the courts into working within a structure that has not functioned as anticipated by the Supreme Court and can fairly be said to not have functioned well at all. 2 Michael H. Graham, Handbook of Federal Evidence, § 702.5, pp. 22-26 (Supp.1998). Graham strongly advises against a rigid application of the Daubert factors and suggests that:
Until the Daubert box is removed, on balance, it is suggested that Daubert’s gatek-eeping language should be held by lower courts to apply to “scientific” evidence only. This interpretation is most consistent with the plain meaning of the opinion and. the clear choice for liberalization if liberal admissibility is in fact the goal. Most importantly, nonapplication of judicial gatekeeping to “technical or other specialized knowledge” would prevent the hardship incurred by many plaintiffs in product liability litigation. Such an interpretation also avoids unthinking application of the four Daubert factors as well as the alternative trying process of developing a list of factors for determining whether a construction worker with 30 years of reinforced concrete experience is testifying to an explanative theory that is sufficiently trustworthy. Id. at 25-26.
In Daubert, the Supreme Court stated: “The inquiry envisioned by Rule 702 is, we emphasize, a flexible one.” Daubert, 509 U.S. at 594, 113 S.Ct. 2786. The en banc majority opinion, however, heedless of Dau-bert’s precept, and unmindful of the other circuits’ unanimous adoption of a flexible approach in applying the Daubert factors, holds that district courts in this circuit must unthinkingly and rigidly apply the Daubert factors in assessing the reliability of a clinical physician’s opinion as to the causal relationship between an individual’s exposure to a chemical or substance and that person’s disease or medical disorder.4 This means, of *286course, that in cases such as the present one, in which the association between a specific chemical compound and a particular disease has not yet been, and perhaps never will be, subjected to hard science investigation, that the plaintiff will be unable to present any expert testimony that his or her exposure to the chemical compound was the probable medical cause of his or her disease.
The en banc majority adopts a mechanistic interpretation of the Daubert factors that threatens to require the exclusion from evidence of vast numbers of clinical medical opinions, although they are generally accepted as trustworthy by physicians practicing in their fields, and, until the majority’s decision today, were routinely accepted as reliable by our courts both before and after Daubert. See Carroll v. Morgan, 17 F.3d 787, 789-90 (5th Cir.1994). Disturbingly, the majority does not explain the reasons for its deviation from the other circuits or its departure from the prior precedent and practice in our courts. Ironically, the majority’s divergence occurs in a rather run-of-the-mill setting, a ease involving a clinical physician’s opinion, based on generally accepted clinical methodology, as to the cause of a non-catastrophic disease following a person’s episodic and traumatic occupational exposure to a chemical compound. Unlike Daubert, and other highly publicized toxic torts cases, the present case does not involve “junk science,” or purportedly hard scientific opinions, based on epidemiological and animal studies not generally accepted in their discipline, as to the surreptitious causal relationship between drugs or other substances and catastrophic systemic diseases or disorders such as cancer and birth defects.
(b)
Having depleted the ranks of medical causation experts available to plaintiffs suffering non-catastrophic chemical exposure injuries, the majority adds insult to injury by casting doubt on the importance of a principal element used by both hard scientific and clinical medical experts in determining whether there is a causal relationship between an individual’s exposure to a substance and his or her disease viz., the temporal relationship between the person’s exposure and the development of symptoms or signs of disease. The majority asserts that in the absence of an established scientific connection between exposure and illness or compelling circumstances, the temporal connection between exposure to chemicals and an onset of symptoms is entitled to little weight in determining causation. Maj. Op. at p. 278. This dictum conflicts with the great weight of scientific and judicial authority.
In the sphere of hard science, the opinion of an expert who opines that exposure to a compound caused a person’s disease is “based on an assessment of the individual’s exposure, including the amount, the temporal relationship between the exposure and disease, and exposure to other disease-causing factors.” Federal Judicial Center, REFERENCE Manual on Scientifio Evidence, p. 205 (1994)(emphasis added). The temporal relationship may either support or contradict causation. “In most acute injuries, there is a short time period between cause and effect. However, in some situations, the length of basic biological processes necessitates a longer period of time between initial exposure and the onset of observable disease.” Id. at 207. Moreover, temporal relationship is one of the seven factors that an epidemiologist considers in determining whether the association between an agent and a disease is causal. Id. at 161.
Courts and commentators have also recognized that the fact that an individual’s symptoms followed an appropriate time after exposure is an important consideration in determining causation. E.g., Kannankeril v. Terminix Int’l., Inc., 128 F.3d 802, 805, 809 (3rd Cir.1997); Zuchowicz, 140 F.3d at 385 (affirming the admissibility of an expert whose “conclusion was based on the temporal relationship between the overdose and the start of disease and the differential etiology method of excluding other possible causes.”); 1 Margie Searcy-Alford, A Guide to Toxic Torts § 10.03[2], p. 10-69 (1998)(“The fact that the symptoms follow an appropriate time after exposure does not prove causation, but it is an important consideration.”); Stephen A. Saltzburg et al, Federal Rules of Evidenoe Manual at *2871233-1234 (7th ed.1998); see Benedi v. McNeil-P.P.C., Inc., 66 F.3d 1378, 1384 (4th Cir.1995); 3 Stuart M. Speiser et al., The Ameuican Law of Torts § 11.27, at 465 (1986).
The district court ease relied on by the majority, Cavallo v. Star Enter., 892 F.Supp. 756 (E.D.Va.1995), is distinguishable in numerous respects and does not support the majority’s assertion that temporal relationship is entitled to “little weight” in the absence of compelling circumstances. In Ca-vallo, the plaintiffs exposure occurred in the open parking lot of a shopping mall during a five minute period at a distance of 500 feet from the source of the jet fuel fumes, the chemical substance at issue; she did not seek medical assistance until nine days later for her symptoms that resulted in an initial diagnosis of “conjunctivitis, or eye redness;” her experts did not have even a rough idea of the amount of her exposure; and there was no showing that the fumes the plaintiff inhaled from the defendant’s alleged negligent spillage were actually more dense than the ordinary daily atmosphere in the shopping mall near defendant’s petroleum distribution, mixing and transfer terminal. Significantly, Ca-vallo’s experts did not have a material safety data sheet (MSDS) or full knowledge of some of the chemicals inhaled and, more importantly, they did not reliably use or apply the methodology of their own disciplines.
In sum, the Cavallo court ruled the experts’ opinions inadmissible because their opinions were based almost exclusively on a very tenuous temporal and spatial connection between exposure and symptoms and because they significantly departed from the accepted toxicology methodology, while the defendant’s toxicology expert followed the generally accepted methodology of that discipline. Id. at 763, 773. Moreover, the Caval-lo court never said that, in the absence of compelling circumstances, a temporal relationship is “entitled to little weight.” Instead, that court merely observed that there may be instances where the temporal connection is so compelling as to dispense with the need for toxicologists to rely on the standard methodology of them discipline. Id. at 773.
(c)
As a coup de grace to inhalation injury claimants, the majority indicates that, if a plaintiffs expert does not have scientifically accurate measurements of the level of the plaintiffs exposure, “his causation opinion [will be] suspect even if he ha[s] scientific support for the position that the [chemical compound] could cause [the plaintiffs disease].” Maj. Op. at p. 278 n. 9. The majority downplays the lethal swath of its new rule by suggesting that it applies here because of “the paucity of the facts Dr. Jenkins had available about the level of Mr. Moore’s exposure.” But the truth is that Dr. Jenkins had better information about the nature of the substances, the level of exposure, and its duration than experts in most inhalation accident cases.5 “Only rarely are humans exposed to chemicals in a manner that permits a quantitative determination of adverse outcomes. [ ] Human exposure occurs most frequently in occupational settings where workers are exposed to industrial chemicals like *288lead or ásbestos; however, even under these circumstances, it is usually difficult, if not impossible, to quantify the amount of exposure.” Federal Judicial Center, REFERENCE Manual on SCIENTIFIC Evidenoe, p. 187 (1994). Consequently, the majority’s rule will apply in virtually all inhalation cases to exclude the opinions of plaintiffs’ experts as to specific medical causation even if they are fortunate enough to have hard science data supporting a general causal relationship or association between the chemical compound and the disease involved. The majority does not have even a paucity of authority to support this extra, gratuitous ratcheting down of inhalation accident victims’ chances of recovery.
2.
The majority has conducted a trial de novo of the district court’s preliminary assessment of whether the reasoning and methodology underlying Dr. Jenkins’ testimony was reliable, substituting its own erroneous judgment and reasoning for that of the trial judge, rather than reviewing the district court’s rulings and reasoning for abuse of discretion, General Electric Co. v. Joiner, - U.S. -, 118 S.Ct. 512, 517, 139 L.Ed.2d 508 (1997), clearly erroneous factual findings, Bourjaily v. United States, 483 U.S. 171, 181, 107 S.Ct. 2775, 97 L.Ed.2d 144 (1987), and errors of law, Koon v. United States, 518 U.S. 81, 100, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996)(“A district court by definition abuses its discretion when it makes an error of law”).
In the district court proceedings, the defendants objected to the introduction of Dr. Jenkins’ opinion as to the diagnosis and cause of Mr. Moore’s disease on the grounds that the doctor lacked hard scientific support that the chemical compound involved could cause reactive airways disease. The district court admitted Dr. Jenkins’ opinion that Mr. Moore had reactive airways disease but excluded Dr. Jenkins’ opinion that the disease had been specifically caused by exposure to the chemical compound involved because Dr. Jenkins had not presented any hard scientific support for a general causal link or association between that particular compound and that particular disease.6
The majority opinion retries the preliminary assessment of Dr. Jenkins’ proffer de novo and concludes that (1) the district court was “entitled to conclude” that (a) Dr. Jenkins had not explained in sufficient detail how his differential diagnosis or etiology and his training and experience were helpful in reaching his conclusion on causation; (b) the MSDS had limited value in supporting Dr. Jenkins’ opinion because he did not know what tests Dow had conducted in preparing the MSDS or what level of exposure was necessary for a person to sustain the injuries warned of in the MSDS; (c) Mr. Moore’s asthma in his youth, history of smoking and recovery from pneumonia shortly before his exposure made Dr. Jenkins’ opinion even more unreliable; and (d) the “analytical gap” between Dr. Jenkins’s causation opinion and the scientific knowledge and available data advanced to support that opinion was too wide; and (2) Dr. Jenkins did not explain precisely how the irritating properties in the compound described by the MSDS were similar to those in other chemicals or compounds that had been linked with reactive airways disease.
Dr. Jenkins testified that he did not know what tests Dow had performed in preparing the MSDS warnings of the hazards of the chemical compound. The district court commented on this fact but based its ruling on the lack of hard scientific support for the doctor’s clinical medical opinion, not on his *289lack of knowledge of Dow’s testing. The MSDS was introduced without objection and referred to in testimony by the experts on both sides, none of whom professed to have any knowledge of Dow’s MSDS-related testing. The record clearly demonstrates that Dr. Jenkins used the MSDS only for the same purpose as did the other experts, merely as a source of information as to the kinds of chemicals in the compound to which Mr. Moore had been exposed. Thus, the district court evidently gave no weight to the experts’ lack of knowledge of Dow’s testing, and if it did find any relevance in this fact, it would have been clearly erroneous in doing so. See Moore, 126 F.3d at 701.
The district court, moreover, did not base its decision on many of the findings and reasons that the majority now attributes to it. Neither the defendant nor the district court found any fault with Dr. Jenkins’ qualifications7, experience, testimony regarding the similarity of irritating chemical properties, or his proper performance of differential etiology to eliminate alternative causes of Mr. Moore’s disease. Because the defendant did not object to Dr. Jenkins’ opinion on these grounds or question him on these points and the district court did not base its ruling on them, these issues should not be raised sua sponte by this court. The performance of physical examinations, taking of medical histories, and employment of reliable laboratory tests provide significant evidence of a reliable differential diagnosis and prima facie evidence that a doctor has considered alternative causes and has attempted to test his or her initial hypothesis as to cause. See Paoli 35 F.3d at 759. The failure of the defendant or the district court to ask for, or the doctor’s failure to volunteer, further elaboration on how each differential diagnosis test is designed to eliminate each alternative cause of disease or a chemistry professor’s exegesis on the structure and composition of each chemical identified as having similar irritating properties, does not afford a proper basis for an appellate trial de novo on the record of the district court’s preliminary assessment hearing.
Likewise, the defendants did not contend, and the trial judge did not rale, that Dr. Jenkins’ opinion was inadmissible because of Mr. Moore’s childhood asthma, smoking or pneumonia. Dr. Jenkins concluded that the exposure to the chemical compound triggered Mr. Moore’s reactive airways disease after taking these and all other relevant factors into consideration. The plaintiff is not required to prove that the exposure was the exclusive cause of the disease. It is well settled in Texas and elsewhere that a defendant takes the plaintiff as he finds him. Coates v. Whittington, 758 S.W.2d 749, 752 (Tex.1988)(citing Driess v. Friederick, 73 Tex. 460, 11 S.W. 493, 494 (1889)); Mondragon v. Austin, 954 S.W.2d 191, 194 (Tex.Ct.App.1997); see Maurer v. United States, 668 F.2d 98, 99-100 (2nd Cir.1981)(“It is a settled principle of tort law that when a defendant’s wrongful act causes injury, he is fully liable for the resulting damage even though the injured plaintiff had a preexisting condition that made the consequences of the wrongful act more severe than they would have been for a normal victim. The defendant takes the plaintiff as he finds him.”); W. Page Keeton, et al., Prosser and Keeton on Torts § 43 at 291-92 (5th ed.1984).
The majority’s most blatant addition of its own ex post facto finding and rationale in an effort to bolster the district court’s ruling, however, is its erroneous claim that the dis-*290triet court found “that the ‘analytical gap’ between Dr. Jenkins’s causation opinion and the scientific knowledge and available data advanced to support that opinion was too wide.” Maj. Op. p. 279. The district court made no such finding. The term “analytical gap,” comes from the Supreme Court’s Joiner opinion of 1997, see 118 S.Ct. at 519, and does not appear in the district court’s 1995 ruling in the present case.8 Moreover, as explained above, the district court based its decision on the same erroneous theory as the majority’s primary rationale, i.e., that a clinical medical physician cannot express an admissible opinion, regardless of how soundly he or she relies on and applies well settled clinical medical methodology, unless the opinion is further supported by hard science, rigid Daubert factor type data.
Conclusion
In the final analysis, this case presents the legal question of the proper interpretation of Federal Rule of Evidence 702 and Daubert in cases involving expert witness proffers based on knowledge beyond the realm of hard scientific knowledge. Indeed, the majority en banc opinion is far too “rulefied” for anyone to seriously contend that it does not set broad, eccentric precedents that will profoundly affect the trials and outcomes in substantial numbers of future cases involving injuries and diseases alleged to have been caused by exposure to chemical compounds. The en banc majority, in my opinion, makes several errors of law, the most serious of which is its holding that a clinical medical expert, whose opinion is based on a sound application of the principles and methodology of his or her discipline, cannot reliably testify as to the causal relationship between and individual’s exposure to a chemical compound and his or her subsequent onset of symptoms and disease. As a result of this error of law and others, the en banc opinion subverts the liberal thrust of the Federal Rules of Evidence and the principles enunciated in Dau-bert by locking the gate on causation evidence derived through the principles and methodology of clinical medicine.
. Evidently, the majority interprets the final Dau-bert factor, "general acceptance,” to mean acceptance within a relevant "hard scientific” community. For it is undisputed that the methods and techniques used by Dr. Daniel Jenkins to determine that Mr. Moore's RADS had been caused by his exposure to the chemical compound, i.e. history taking, physical examinations, differential etiology (conducting tests to eliminate other diagnoses and causes of the patient's disease), and review of other physicians' reports were generally accepted within the doctor’s own clinical medical disciplines of pulmonary and environmental medicine.
. In Daubert, the Court stated:
Scientific conclusions arc subject to perpetual revision. Law, on the other hand, must resolve disputes finally and quickly. The scientific project is advanced by broad and wide-ranging consideration of a multitude of hypotheses, for those that are correct will eventually be shown to be so, and that in itself is an advance. Conjectures that are probably wrong are of little use, however, in the project of reaching a quick, final and binding legal judgment — often of great consequence — about a particular set of events in the past. We recognize that, in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations. Daubert, 509 U.S. at 597, 113 S.Ct. 2786.
The majority en banc opinion quotes this passage at page 275-276 and proceeds to stand it on its head on page 276, interpreting the Supreme Court’s words as supporting the majority's proposition that although hard scientific proof of medical causation will not always be available in chemical injury cases, the cases must be quickly resolved; therefore, in chemical injury cases, if the plaintiff can produce only clinical medical experts whose opinions are based solely on well accepted clinical medicine methodology, they must face trial without a medical causation expert witness.
The Daubert Court neither expressed nor implied such a draconian rule. Being confronted with a case involving the admissibility of hard science epidemiological expert opinions, not generally accepted in that field, proffered to prove that Bendectin could have caused birth defects in children whose mothers used the drug, the Court concluded that the evidence could not be excluded under the Frye rule which was superseded by the Federal Rules of Evidence, but that the trial judge as gatekeeper must determine that the hard science evidence proffered is not only relevant but also reliable as based on a sound application of the methodology of the expert's discipline and suggested several ways, based on basic elements of hard science methodology, that a party who proffers an expert who proposes to testify to a hard scientific opinion can show that the opinion is reliable or, reciprocally, that a court can use to test the opinion’s reliability.
Those ways of testing or showing reliability of hard scientific opinions have become known as the "Daubert factors.” But the Court did not intend to require that these gauges of reliability be applied monolithically to all expert testimony. When the expert does not propose to testify to an opinion based on hard scientific methodology, the Court indicated that the reliability of his opinion should be assessed according to the methodology of the expert’s own discipline. The Daubert court did not indicate, and this court is not called upon to decide, what a trial court should do if it is confronted by proffers of experts who propose to testify to directly conflicting opinions as to medical causation, one based on hard scientific methodology and the other based on clinical medical methodology. In such a case, it is likely that the trial court should find the clinical medical expert’s opinion unreliable if it fails to take into account and distinguish the hard scientific expert’s opinion and its basis in hard scientific data, if the court finds the latter to be reliable. The Daubert Court did no suggest, however, that the Federal Rules of Evidence authorize a federal court to formulate a rule, as the en banc majority'has done, that, in effect, bars a clinical physician from expressing an opinion as to the probable chemical causation of a disease in a specific individual until the existence of a general causal relationship has been confirmed by the use of hard scientific methodology.
. In addition to the views expressed by commentators and practitioners, Stephen A. Saltzburg, et al. 2 Federal Rules of Evidence Manual at 1250-1251 (7th ed.1998) reports that:
The Advisory Committee on Evidence Rules has made a determination that Rule 702 should be amended in light of Daubert and its progeny. The Advisory Committee has prepared a working draft for an amended 702, which, at this writing, has yet to receive final approval from the Committee. The working draft, which is adapted from a proposal by Professor Michael Graham, reads as follows:
Testimony providing scientific, technical or other specialized information, in the form of an opinion, or otherwise, may be permitted if:
(1)the information is based upon adequate underlying facts, data or opinions;
(2)the information is based upon a methodology either (a) established to have gained widespread acceptance in the particular field to which the explanative theory belongs, or (b) shown to possess indicia of trustworthiness;
(3)the methodology has been applied reliably to the facts of the case;
(4) the witness is qualified as an expert by knowledge, skill, experience, training or education to provide such information; and
(5) the information will assist the trier of fact to understand the evidence or to determine a fact in issue.
While the language set forth above is still in development, the Advisory Committee has agreed upon some general substantive points. First, the gatekeeper standards of Rule 702 must apply to all expert testimony. Sec-
ond, the reliability standards must apply not only to the theory or methodology used by the expert, but also to the application of that theory or methodology in the specific case.... Third, it does not pay to get too detailed about the factors that a Trial Judge should use in assessing reliability.... The risk of leaving out important reliability factors is especially great because experts in different fields will necessarily use different methodologies, and it would be very difficult to describe an all-inclusive list of reliability factors that would cover the testimony of all experts.
. The panel opinion in the present case, Moore v. Ashland Chemical, Inc., 126 F.3d 679 (5th Cir.1997), consistently with the foregoing authorities, concluded that: (1) the basic principles of the Federal Rules of Evidence recognized in Daubert apply to the admission or exclusion of every type of expert testimony; (2) a trial judge, therefore, must assess every proffer of expert testimony to determine whether it is relevant to the case and a reliable application of the principles and methodology of that expert’s discipline; (3) the Supreme Court in Daubert interpreted "scientific knowledge" under Federal Rule of Evidence 702, for purposes of that case, to mean knowledge obtained and tested by the scientific method, i.e., "hard” scientific knowledge; (4) accordingly, the Daubert court indicated that a trial court should assess the reliability of expert testimony professedly based on "hard" scientific knowledge using several factors, the "Daubert factors,” which are "hard” science methods or techniques; (5) clinical medicine (as opposed to research and laboratory medical science) is not, strictly speaking, a "hard” scientific discipline; its goals, subject matter, conditions of study, and well developed, sui generis methodology are quite different from that of purely "hard” science and its methodology; (6) Consequently, a trial judge assessing the reliability of the proffer of a clinical physician's experl testimony based on clinical medical knowledge, without purporting to be based on hard scientific methodology, should determine whether it is a sound application of the knowledge, principles and methodology of clinical medicine; (7) In the present case, the district court committed an error of law by rigidly applying the "Daubert factors” and excluding the expert clinical physician's opinion because the doctor did not have any "hard" scientific data to support his clinical medical opinion.
. As explained by the panel opinion:
From Moore's history that Dr. Jenkins had taken, he had information that before the exposure Moore was in good health, that two 400 pound drums of the chemicals had begun leaking in the back of Moore’s truck at some time before his arrival at Ashland, that Moore's rig consisted of a diesel tractor and a 28 foot enclosed trailer, that after the discovery of the leakage upon arrival at Ashland the drums were allowed to continue to leak inside the trailer with the doors shut for another 45 minutes until the Ashland supervisor told Moore to remove them, that at this point the 400 pound drums had become light enough to allow Moore and others to roll them manually out onto the dock, that Moore and a co-employee worked in and around the trailer for about 45 to 60 minutes sprinkling "Absorbo” over the contaminated areas sweeping the saturated material into shovels, removing the materials from the trailer, and shoving the leaking drums into salvage drums, that Moore finished the cleanup at Ashland about 11:00 a.m., that Moore began to experience tightness of chest at about 11:45 a.m., that as his symptoms were continuing to worsen Moore consulted the company doctor who put him on "oxygen and inhalants.” Moore, 126 F.3d at 702.
From this information, Dr. Jenkins was able to roughly estimate that Mr. Moore had been exposed to possibly "200 parts per million or higher” of the chemical compound. Id. at 695.
. Dr. Jenkins performed a physical examination, took a detailed medical history, observed Moore on three occasions, reviewed the MSDS prepared by Dow Corning, and performed a series of tests on Moore including pulmonary function tests, a bronchodilator test, a spirometry test, a plethys-mographic test, a lung volume determination, an intrapial gas distribution test, a diffusion test, an arterial bloods test, a mechanics test, X-rays, and laboratory tests. Dr. Jenkins reviewed the medical records and reports of a bronchodilator test performed by Dr. Simi two to three weeks after the accident that showed severe airways obstruction. Additionally, Dr. Jenkins reviewed a report of an allergy test performed by Dr. Alvarez, which ruled out allergic or immunologic disease and confirmed RADS. Finally, Dr. Jenkins also relied upon the temporal proximity between the exposure to the chemicals at the Ashland facility and the onset of symptoms.
. The majority opinion fails to point out that Dr. Jenkins' qualifications were never an issue at any point in these proceedings. In fact, Dr. Jenkins was more than eminently qualified to render an opinion in this matter as a brief summary of his education, training and experience reveals. Dr. Jenkins received his medical degree from the University of Texas in 1940, received training at the University of Michigan Hospital as an intern, resident in Tuberculosis and Chest Disease and resident in Allergy in 1940-45, served as Instructor and Chief Resident in Medicine and Assistant of Medicine and Physician in charge of the Tuberculosis and Chest Unit, University of Michigan Medical School, 1943 to 1947, was certified by the American Board of Internal Medicine in 1947, served in various capacities as a professor at Baylor College of Medicine from 1947-91 where from 1947-74 he was chief of the Pulmonary Disease Section and from 1975-91 chief of environmental medicine. Additionally, in the course of over fifty years of practicing medicine. Dr. Jenkins has examined and evaluated over 100 persons for injuries occurring from exposure to various chemical compounds in an occupational setting.
. In General Electric Co. v. Joiner, - U.S.-, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997), the Supreme Court held that abuse of discretion, rather than the particularly stringent standard of review applied by the court of appeals in that case, is the proper standard by which to review a district court’s decision to admit or exclude scientific evidence. The plaintiff Joiner proffered expert testimony based on hard science methodology, animal and epidemiological studies, to prove that the defendants' PCBs and related products had caused his lung cancer. "Joiner’s experts used a 'weight of the evidence’ methodology to assess whether Joiner’s exposure to transformer fluids promoted his lung cancer. They did not suggest that any one study provided adequate support for their conclusions, but instead relied on all the studies taken together (along with their interviews of Joiner and their review of his medical records).” Id., 118 S.Ct. at 521 (Stevens, J. concurring in part and dissenting in part) (footnote omitted). The district court examined the studies and excluded the experts’ opinions on the ground that none of the studies was sufficient alone to show a link between PCBs and lung cancer.
The Supreme Court held that the district court did not abuse its discretion in excluding the experts’ testimony on grounds that the studies upon which the experts relied were not sufficient, whether individually or in combination, to support their conclusions. The Supreme Court remarked that "[a] court may conclude that there is simply too great an analytical gap between the data and the opinion proffered.” Id. at 519.
In the present case, there was no "analytical gap” between Dr. Jenkins’ data and his opinion that Mr. Moore’s exposure caused his disease. In fact, the district court allowed Dr. Alvarez to use the identical data to express the same opinion. It is easy to see that the district court’s decision in Joiner was reasonable and not an abuse of discretion because the plaintiff himself conceded that there was an analytical gap between each one of his expert's studies and the conclusion that PCBs caused his cancer. He argued, although unsuccessfully, however, that every analytical gap could be bridged if all of the experts’ studies were considered in combination. In the present case, the district court excluded Dr. Jenkins’ opinion simply because he did not have any hard scientific support for his clinical medical opinion, not because of a gap in reasoning. Dr. Jenkins’ clinical medical opinion was, in fact, snugly based on the sound application of the well accepted methodology of his discipline. Thus, era banc the majority itself is simply attempting to bridge too great an analytical gap by trying to stretch Joiner to cover the present case.