MANSFIELD, J., delivered the opinion of the Court, in which McCORMICK, P.J., and KELLER, HOLLAND, and KEASLER, JJ., joined. JOHNSON, J., filed a dissenting opinion, in which MEYERS, PRICE,and WOMACK, JJ., joined.

The question presented is whether the Ninth Court of Appeals erred in holding that the trial court abused its discretion in excluding expert testimony on the reliability of eyewitness identifica tions. We hold that the Court of Appeals did err.

A Montgomery County grand jury indicted appellant, Jon David Weatherred, for the capital murder of William Ralph Strawn in The Woodlands in 1988. See Tex. Pen. Code § 19.03(a)(2). At trial, the State’s case relied in part on the testimony of two eyewitnesses who identified appellant as being near the scene of the crime at or about the time of the crime. In rebuttal to those eyewitnesses, appellant offered the testimony of Dr. Kenneth Deffenbacher, a psychologist who claimed to be an expert on the reliability (or unreliability) of eyewitness identifications.²

The trial court held a hearing, outside the presence of the jury, on the admissibility of Deffenbacher’s testimony under Texas Rule of Evidence 702.³ At that hearing, Deffenbacher testified, in relevant part, that (1) he had a doctoral degree in psychology and was chairman of the department of psychology at the University of Nebraska at Omaha; (2) he had done extensive research in the field of human visual perception and memory; (3) he had written “about 35 articles, ten or twelve chapters in edited books, and a textbook” on human visual perception and memory; (4) he and other psychologists had identified, through generally-accepted experimental research, numerous “variables” affecting the reliability of eyewitness identifications; and (5) three of those variables — “photo bias,” the “forgetting of a stranger’s face,” and the lack of a relationship between eyewitness confidence and eyewitness accuracy — were applicable to the eyewitness identifications in the instant case.

At the conclusion of the hearing, the trial court ruled Deffenbacher’s testimony inadmissible. The trial court gave no explanation for its ruling. Five days later, appellant asked, and was allowed, to supplement his offer of proof with photocopies of Deffenbacher’s curriculum vitae and five published articles, some by Deffenbacher and some by others, that reviewed current psychological research on the reliability of eyewitness identifications. Appellant did not, however, ask the trial court to reconsider its earlier ruling. The jury subsequently found appellant guilty as charged in the indictment. Punishment was assessed at imprisonment for life.

On direct appeal, appellant brought six points of error. In one of the points, he argued that the trial court erred in refusing to let Deffenbacher testify before the jury. The State maintained, in response, that the trial court did not abuse its discretion in excluding Deffenbacher’s testimony because “[a]ppellant failed to ... timely present sufficient evidence” of the scientific reliability of that testimony. The Ninth Court of Appeals accepted appellant’s argument and rejected the State’s after “finding,” apparently as a matter of law, that Deffenbacher’s testimony was both relevant and reliable. The court then reversed the judgment of the trial court and remanded the case for a new trial. Weatherred v. State, 985 S.W.2d 234 (Tex.App.--Beaumont 1999). We granted the State Prosecuting Attorney’s petition for discretionary review to determine whether the Court of Appeals erred in accepting appellant’s argument. See Tex. R. App. Proc. 66.3(c).

Under Rule 702, the proponent of scientific evidence must show, by clear and convincing proof, that the evidence he is proffering is sufficiently relevant and reliable to assist the jury in accurately understanding other evidence or in determining a fact in issue.⁴ Nenno v. State, 970 S.W.2d 549, 560-561 (Tex.Crim.App. 1998); Hartman v. State, 946 S.W.2d 60, 62-63 (Tex.Crim. App. 1997); Jordan v. State, 928 S.W.2d 550, 553-555 (Tex.Crim.App. 1996); Kelly v. State, 824 S.W.2d 568, 572-573 (Tex.Crim.App. 1992). The reliability of “soft” science evidence,⁵ such as was offered in the instant case, may be established by showing that (1) the field of expertise involved is a legitimate one, (2) the subject matter of the expert’s testimony is within the scope of that field, and (3) the expert’s testimony properly relies upon or utilizes the principles involved in that field. Nenno v. State, 970 S.W.2d at 561.

An appellate court reviewing a trial court’s ruling on the admissibility of evidence must utilize an abuse-of-discretion standard of review. Prystash v. State, 3 S.W.3d 522, 527 (Tex.Crim.App. 1999). In other words, the appellate court must uphold the trial court’s ruling if it was within the zone of reasonable disagreement. Montgomery v. State, 810 S.W.2d 372, 391 (Tex.Crim.App. 1990). In addition, the appellate court must review the trial court’s ruling in light of what was before the trial court at the time the ruling was made. Hoyos v. State, 982 S.W.2d 419, 422 (Tex.Crim.App. 1998); Hardesty v. State, 667 S.W.2d 130, 133 n. 6 (Tex.Crim.App. 1984).

When appellant proffered Deffenbacher’s expert testimony to the trial court, appellant had the burden of proving by clear and convincing evidence that the testimony was relevant and reliable and not mere “junk science.” Appellant attempted to carry that considerable burden, at that critical time, by simply offering Deffenbacher’s testimony and nothing else.⁶ Furthermore, a close examination of Deffenbacher’s testimony reveals that, although he claimed that he and others had carried out extensive research on the reliability of eyewitness identifications and that he himself had written much on that subject, he failed to produce or even name any of the studies, researchers, or writings in question. The trial court did not state its reason for excluding Deffenbacher’s testimony, but, given what the trial court had before it at the time it ruled, it could have reasonably concluded that appellant failed to carry his burden of showing that the proffered expert testimony was scientifically reliable. See Jordan v. State, 950 S.W.2d 210, 212 (Tex.App.--Fort Worth 1997, pet. ref’d) (record did not show abuse of discretion on part of trial court in excluding expert testimony on reliability of eyewitness identifications when proffered expert witness “failed to mention by name any other person who purports to be an expert in the field or produce or name the studies he relied on to reach his opinions”). Therefore, the Court of Appeals erred in holding that the trial court abused its discretion.

We reverse the judgment of the Court of Appeals and remand the case to that court so that it may consider appellant’s remaining points of error.

DELIVERED MARCH 29, 2000

¹ This case has a complicated procedural history, but we discuss today only that part that is relevant to this appeal. Those wishing more may see Weatherred v. State, 963 S.W.2d 115 (Tex.App.--Beaumont 1998), vacated and remanded, 975 S.W.2d 323 (Tex.Crim.App. 1998), on remand, 985 S.W.2d 234 (Tex.App.--Beaumont 1999).

² The proffer of such evidence has become increasingly common, but it remains controversial in most jurisdictions. See P. Giannelli & E. Imwinkelried, Scientific Evidence § 9-2 (3^rd ed. 1999); J. McLaughlin, et al., Weinstein’s Federal Evidence § 702.03[6] (2^nd ed. 1999); G. Sarno, Annotation, Admissibility, at Criminal Prosecution, of Expert Testimony on Reliability of Eyewitness Testimony, 46 A.L.R. 4th 1047 (1986 & Supp. 1999). See also Jordan v. State, 928 S.W.2d 550 (Tex.Crim.App. 1996).

³ Rule 702 provides that “[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify thereto in the form of an opinion or otherwise.”

⁴ Of course, once a particular type of scientific evidence is well established as reliable, a court may take judicial notice of that fact, thereby relieving the proponent of the burden of producing evidence on that question. Emerson v. State, 880 S.W.2d 759, 764 (Tex.Crim.App.), cert. denied, 115 S.Ct. 323 (1994); J. McLaughlin, et al., Weinstein’s Federal Evidence § 702.05[1] (2^nd ed. 1999); P. Giannelli & E. Imwinkelried, Scientific Evidence § 1-2 (3^rd ed. 1999).

⁵ The “hard” sciences, areas in which precise measurement, calculation, and prediction are generally possible, include mathematics, physical science, earth science, and life science. The “soft” sciences, in contrast, are generally thought to include such fields as psychology, economics, political science, anthropology, and sociology. See The New Columbia Encyclopedia 2450 (1975).

⁶ One scholar has suggested that expert testimony may be shown reliable “by deposition testimony, affidavits, proffers, stipulations, learned treatises, testimony or some combination thereof.” R. Murrian, The Admissibility of Expert Eyewitness Testimony Under the Federal Rules, 29 Cumb. L. Rev. 379, 385 (1999). Accord, United States v. Downing, 753 F.2d 1224, 1241 (3^rd Cir. 1985).