February 22, 1999

No. 97SC758





Certiorari to the Colorado Court of Appeals



David F. Vela, Colorado State Public Defender
Karen N. Taylor, Deputy State Public Defender

Denver, Colorado
Attorneys for Petitioner

Ken Salazar, Attorney General
Barbara McDonnell, Chief Deputy Attorney General
Michael E. McLachlan, Solicitor General
Roger G. Billotte, Assistant Attorney General
Criminal Enforcement Section

Denver, Colorado
Attorneys for Respondent

Wells, Anderson & Race LLC
Mary A. Wells
L. Michael Brooks, Jr.

Denver, Colorado

The Product Liability Advisory Council, Inc.
Hugh F. Young, Jr.

Reston, Virginia
Attorneys for Amicus Curiae, The Product Liability Advisory Council, Inc.

JUSTICE MARTINEZ delivered the Opinion of the Court.

In this case, we consider the admissibility of expert testimony describing the scent tracking and purported identification of a criminal defendant by a trained police bloodhound. Specifically, we determine whether the "general acceptance" standard embodied in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), or the scientific validation factors enumerated by the U.S. Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), are useful to analyze scent tracking evidence.1

We conclude that testimony on this subject is not susceptible to review under standards reserved for "scientific" evidence. Accordingly, we decline to apply the Frye test or the Daubert factors. Instead, we hold that scent tracking evidence must be governed by a conventional CRE 702 and CRE 403 analysis, and we adopt the majority rule for assessing the reliability of such evidence.

The court of appeals found that the scent tracking evidence in this case was properly before the jury. See People v. Brooks, 950 P.2d 649, 653 (Colo. App. 1997). Because our review of the record supports the court's conclusion, we affirm.


Dwayne Brooks, the petitioner and defendant below, was charged with second degree burglary,2 attempted theft,3 and possession of burglary tools4 in connection with events that took place during the early evening hours of November 15, 1993. On the night in question, a resident called the police after he discovered someone breaking into the home of his neighbor.

Within minutes, several police officers arrived at the scene, taking cover some 100 to 120 feet away. The victim's house was illuminated by a nearby streetlight, and the ground was covered with several inches of new snow. After a few minutes, several witnesses saw the suspect exit the house and proceed toward the curb, but none were able to get a good enough look at his face to offer a conclusive identification. An officer aired the following description: a black male with a medium build, wearing dark pants, a gray jacket, and perhaps white gloves. Another officer gave a similar description, indicating a stocky black male wearing dark clothing. The neighbor who had called in the report believed that the burglar was wearing a dark colored ski suit. The height and weight estimations given by these witnesses varied.

The suspect walked away from the house, paused when police called for him to halt, and then ran in the opposite direction. As officers pursued on foot, the suspect darted between two houses and began jumping fences, traveling from backyard to backyard in the residential subdivision. One officer radioed for a "K-9" scent tracking unit and remained at the fence to point out the suspect's footprints in the fresh snow to the K-9 unit, while others followed the tracks.

Officer Jerry Nichols arrived with his dog "Yogi," a purebred bloodhound he had been training since about nine or ten weeks of age. This training consisted of exposing the dog to an object handled by a target individual, and then requiring that Yogi find this person. Positive reinforcement was apparently used to facilitate the process. Thus, when the dog successfully found the target, he was rewarded with a treat. As the training progressed, the trails were made longer and more difficult, with distractions and other irregularities added.

Yogi had been trained to "drop-trail." That is, he had the ability to continue following a scent even after being pulled off the original trail for a period of time. In training exercises, this would be accomplished by stopping the dog in the middle of a track, and waiting fifteen to twenty minutes before directing the dog to continue. According to Nichols, Yogi would successfully resume tracking, following the correct trail when tested in this manner.

Out of 480 training sessions, Yogi had failed to track on fourteen occasions, most of which Officer Nichols attributed to handler error. Nichols also kept a record of Yogi's performance in "real-life" tracks. Out of some 390 such deployments, Nichols reported confirmed finds by Yogi in 214 cases. A find was considered "confirmed" if it was supported by independent corroborating evidence.

Yogi was led to the footprints left by the suspect in this case, and the dog began following the trail. During the chase, when Yogi would reach a fence, Officer Nichols would pull him from the trail, move the dog to the other side, and allow him to continue tracking.5 As the officers and Yogi approached one such fence, a man fitting the general description of the suspect aired earlier jumped from some shrubbery on the other side and ran.

Eventually, Yogi led Nichols to a garage attached to another private residence in the subdivision. The other officers who had been following the suspect's footprints arrived at this location as well. While officers did not immediately see anyone, they noticed Yogi sniffing near a pickup truck parked inside. An officer used a flashlight to illuminate the floor under the truck and discovered the defendant Dwayne Brooks hiding beneath the vehicle. Brooks was wearing a gray jacket like that reportedly seen on the suspect, and he fit the general description aired by officers earlier. The jacket Brooks wore was also consistent with the description of the individual flushed from the shadows during trailing efforts.

Brooks struggled with the officers as they attempted to take him into custody. He also refused to provide any personal information, supplying the alias "Mark Scott" instead of his name. During a frisk for weapons, officers discovered two large screw-drivers in the jacket Brooks was wearing, and a Glade "plug-in" air freshener was found in his pocket. The air-fresher was identified as having been taken from the victim's residence, as were a gold ring found on the floor where Brooks was found hiding and a necklace subsequently discovered on his person. The screw-drivers were consistent with pry-marks found on the door of the victim's home.

Once Brooks was under control, the officers attempted to confirm whether he was the individual Yogi had been trailing. Officer Nichols led Yogi back into the garage where Brooks was standing handcuffed among several police officers. Yogi approached Brooks, nuzzling and pawing him, while slobbering and apparently anticipating a treat. According to Nichols, this was consistent with Yogi's training. Based on the dog's behavior, Nichols concluded that Brooks was, in fact, the person whose trail Yogi had been following.

Following arrest and the filing of charges, Brooks entered a plea of not guilty and submitted various pleadings, including a motion to suppress "dog evidence." Specifically, the defense sought to exclude any reference whatsoever to Yogi's involvement in the case, especially the purported identification of Brooks as the source of the scent being tracked.

The defendant maintained that scent tracking should be viewed as scientific evidence and excluded under either Frye v. United States, 293 F. 1013 (D.C. Cir. 1923) or Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). Following an extensive hearing the trial court found that Frye did not apply. Relying on Daubert, however, the court stated that testimony about Yogi's training and performance could only be considered "specialized knowledge . . . in the loosest sense of the term," and would not support the "gloss of expertise" under CRE 702 because no "scientific method" had been offered to explain how dogs track a particular scent. Although it concluded that the evidence would not be allowed as expert testimony under CRE 702, the court qualified its ruling by finding that Officer Nichols might be permitted to testify about the scent tracking and his observations of Yogi as a lay witness pursuant to CRE 701. This question was left for consideration by a visiting judge who would actually be conducting the trial in Brooks' case.

The second judge disagreed with the earlier court's reliance on Daubert, and made a finding at a subsequent hearing that the testimony of Officer Nichols was admissible as expert testimony under CRE 702. However, in deference to the earlier ruling, the second judge decided to admit the evidence only as a lay opinion under CRE 701.

At trial, Nichols discussed Yogi's training and qualifications as detailed above, and related the events leading up to Brooks' arrest on the night of the crime. He gave his opinion that Yogi had identified Brooks as the person who had left the tracks attributed to the perpetrator. Nichols also spoke briefly regarding his own "theory" about bloodhound tracking, relating his view that the dogs can pick up scents left in the form of skin cells shed by humans as they move. He conceded, however, that no one had been able to establish precisely how the dogs were able to track.

Brooks was convicted on all counts, and sentenced to a term of fifteen years in the Department of Corrections. He challenged his convictions before the court of appeals, arguing that the admission of information related to Yogi's scent tracking efforts constituted reversible error. See Brooks, 950 P.2d at 651. The court of appeals was persuaded that, although the evidence had been admitted as lay opinion testimony under CRE 701, it was more properly considered expert testimony subject to CRE 702. See id. at 652. Finding that neither Frye nor Daubert applied, the court concluded that a sufficient foundation had been established to admit the testimony under CRE 702, and it upheld Brooks' conviction. See id. We now affirm.


We have not previously addressed the admissibility of canine scent tracking evidence in the form of expert testimony by a dog's handler. Because this case requires us to choose among several potentially overlapping standards, we define the scope and applicability of each. We begin by discussing the general framework that has governed the admissibility of expert testimony in Colorado, first under the common law and more recently pursuant to CRE 702. We next consider the applicability of the Frye test and the Daubert factors, taking into account the origins and development of each standard and the type of evidence to which each has been applied. We conclude that the particular tests for validity found in Frye and Daubert are not helpful to analyze scent tracking evidence, and that validity and reliability of scent tracking evidence is better ensured by requiring a proper foundation consistent with CRE 702 and CRE 403. Further, we adopt the foundational considerations applied to scent tracking evidence in the majority of other jurisdictions, and we approve of the trial court's reliance on such matters as part of its analysis under CRE 702 and CRE 403 in this case.


The Colorado Rules of Evidence were adopted by this court on October 23, 1979, and became effective on January 1, 1980. Prior to the adoption of the rules, expert testimony in this state was governed by the common law. Even then, two concerns were often apparent. The first was whether the subject matter of an expert's testimony would be useful to the jury. See McNelley v. Smith, 149 Colo. 177, 180, 368 P.2d 555, 557 (1962). The second was whether the particular witness on the stand actually possessed the special skill or knowledge necessary to qualify as an expert on the subject in question. See, e.g., City of Pueblo v. Ratliff, 137 Colo. 468, 475, 327 P.2d 270, 274 (1958); Ausmus v. People, 47 Colo. 167, 188, 107 P. 204, 211 (1909)(defining expert as one with superior knowledge "by professional, scientific, or technical training or by practical experience in some field of human activity"). As early as 1919, this court approved of the following statement of the common law rule:

The true test of the admissibility of such testimony is . . . whether the witnesses offered as experts have any peculiar knowledge or experience, not common to the world, which renders their opinions founded on such knowledge or experience any aid to the court or the jury in determining the questions at issue.

National Fuel Co. v. McNulty, 65 Colo. 176, 180, 177 P. 979, 981 (1919)(citations omitted).

The concerns that first arose in the common law found their way into the modern realm by way of CRE 702, which adopts an approach quite similar to that endorsed in McNulty. Specifically, CRE 702 provides:

If 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.

We have recently reiterated our view that CRE 702 calls for a two-tiered analysis. See Huntoon v. TCI Cablevision, Inc., 969 P.2d 681, 689 (Colo. 1998). A court must first consider whether the substance of the proffered testimony will be helpful to the fact finder. See id.; Melville v. Southward, 791 P.2d 383, 387 (Colo. 1990). The court must then decide whether the witness serving as the conduit for such information is competent to render an expert opinion on the subject in question. See Huntoon, 969 P.2d at 689. Taken together, these two prongs ask the familiar question: "On this subject can a jury from this person receive appreciable help?" People v. Williams, 790 P.2d 796, 798 (Colo. 1990)(emphasis in original).

While this basic framework is relatively straightforward, the courts have moved, under certain circumstances, to independently scrutinize the underlying substance of a proffered expert opinion for notions of reliability and validity. As will be seen, the particularized tests that Brooks would have us apply to the canine scent tracking evidence here are examples of such efforts. Brooks insists that this case implicates "scientific" evidence of the sort governed by the Frye "general acceptance" test, or in the alternative, the validation factors enumerated by the U.S. Supreme Court in Daubert.

In response, the People suggest that before either of these standards can apply, a court must be satisfied that the evidence being scrutinized embodies the sort of "scientific" knowledge visited in those cases. We agree. Because Frye and Daubert are themselves divergent, we consider each separately.


The Frye test takes its name from the 1923 federal circuit court decision in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). That case considered the admissibility of a systolic blood pressure deception test, a predecessor to the modern polygraph ("lie detector"). See id. The Frye court concluded in a brief opinion that evidence pertaining to the procedure was not appropriate in judicial proceedings because the test was not generally accepted as valid in its field. See id. The court explained:

Just when a scientific principle or discovery crosses the line between the experimental and demonstrable stages is difficult to define. Somewhere in this twilight zone the evidential force of the principle must be recognized, and while courts will go a long way in admitting expert testimony deducted from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs.

Id. at 1014 (emphasis added).

When we considered the admissibility of a more modern version of the lie detector test in 1981, we applied Frye to exclude such evidence. See People v. Anderson, 637 P.2d 354, 358 (Colo. 1981). Because Anderson, like Frye, pondered the admissibility of polygraph evidence, it was originally unclear whether we were adopting the "general acceptance" standard for all novel scientific evidence, or only as to the polygraph - a device most courts viewed with suspicion. Indeed, in reaching our decision, we noted that "the polygraph is unlike other scientific evidence, since what it attempts to measure - the truthfulness of a witness or defendant - is so directly related to the essence of the trial process." Anderson, 637 P.2d at 361.

On several occasions subsequent to Anderson, we moved to limit the applicability of Frye. In People v. Romero, we adopted a case by case approach for considering the reliability, and thus, the admissibility of testimony from witnesses who had been hypnotized. 745 P.2d 1003, 1016 (Colo. 1987) Our decision expressly overruled an earlier court of appeals opinion, People v. Quintanar, 659 P.2d 710 (Colo. App. 1982), that had relied on the Frye test to adopt a per se rule excluding such evidence. See Romero, 745 P.2d at 1015 n.8.

In People v. Hampton, 746 P.2d 947, 950 (Colo. 1987), we took up the issue more directly, concluding that CRE 702, and not Frye, governed the admissibility of expert testimony on developing theories pertaining to "rape trauma syndrome." We noted that the specialized review epitomized by Frye was proper only in a particular type of case:

Generally, the Frye test is applied to novel scientific devices and processes involving the manipulation of physical evidence including lie detectors, experimental systems of blood typing, voiceprints, identification of human bite marks, and microscopic analysis of gunshot residue.

746 P.2d at 950-51 (emphasis added).

Finally, in Campbell v. People, we held a trial court's application of Frye to expert testimony on the reliability of eyewitness identification was erroneous, and concluded that the more liberal standard contained in CRE 702 should have been used. 814 P.2d 1, 7 (Colo. 1991). We noted that Frye was essentially a "special foundation requirement for novel scientific devices or processes involving the evaluation of physical evidence," and found no such scientific device or process in the context of "an explanation by a psychologist on how certain factors, such as stress and post-event information, can affect . . . eyewitness identification." Id. at 8 (emphasis added). Because the evidence being considered in Campbell nevertheless raised concerns relative to reliability, we endorsed a case by case approach that focused analysis on both the admissibility standards contained in CRE 702 and the safeguards in CRE 403, thereby "ensuring that the admission of expert testimony [on the subject] will enhance the truth-seeking function of trial." Id.

Although we declined to apply Frye in several instances, we did not abandon the rule altogether. Its general acceptance test was applied once again in Fishback v. People, 851 P.2d 884 (Colo. 1993). There, we concluded that because genetic analysis of deoxyribonucleic acid ("DNA") utilized "a number of highly technical and sophisticated techniques in order to extract, isolate, and observe alleles contained in human DNA molecules," it was "precisely the sort of scientific evidence which requires application of the Frye test." Id. at 890. Moreover, we found the expertise in question to be a relatively new field, observing that "because the potential of DNA typing technology for forensic purposes was first recognized in the mid-1980's, first applied in the late 1980's, and involves techniques which are continuously evolving, DNA typing is a 'novel' scientific process." Id. Writing for the majority, then Chief Justice Rovira explained that requiring novel scientific evidence to be generally accepted in the relevant scientific community is a way to "ensure that only reliable evidence [is] admitted." See id. Accordingly, the DNA evidence in Fishback was analyzed under Frye. Two years later, we again applied a Frye analysis to a dispute concerning DNA evidence. See Lindsey v. People, 892 P.2d 281 (Colo. 1995).

In light of this background, we turn now to Brooks' suggestion that we apply Frye to the canine scent tracking evidence at issue in the instant case. We conclude that even if some aspects of the tracking efforts here could be considered "novel,"6 the Frye test would not be appropriate. In our view, the foregoing discussion demonstrates that the "general acceptance" standard cannot apply unless the expertise being considered is based on, or derivative of, hard science. Indeed, we have applied the Frye test in exactly three cases - one involving the complex scientific trappings of a polygraph test, and two considering the sophisticated scientific theories and procedures surrounding DNA analysis. See Anderson, 637 P.2d at 356 (describing polygraph technique and procedure); Fishback, 851 P.2d at 885 (describing DNA theory and scientific background for restriction fragment length polymorphism "RFLP" analysis, gel electrophoresis, etc.); Lindsey, 892 P.2d at 285 (discussing RFLP analysis and frequency calculations). Where, on the other hand, the expertise in question is outside the realm of hard science, we have declined to apply Frye - even in circumstances where the substance of the proposed testimony might fairly be categorized as related to one or more of the "social" sciences. See, e.g., Hampton, 746 P.2d at 949 (describing rape trauma syndrome as variant of post-traumatic stress disorder recognized by psychiatric professionals); Campbell, 814 P.2d at 3 (describing psychologists' view on factors potentially affecting eyewitness identification).

Perhaps anticipating this weakness in his position, Brooks characterizes the evidence at issue here as implicating the sort of "Frye science" that would justify the invocation of the "general acceptance" test. In this regard, he attempts to analogize a bloodhound to a polygraph machine or voiceprint analysis - insisting that, like these procedures, scent tracking involves the manipulation of physical evidence with a scientific device or process. Here, it is suggested that the evidence being manipulated consists of scent particles, and that the "scientific device" responsible for said manipulation is none other than Yogi the dog.

We are not persuaded. Instead, we agree with the court of appeals that "[t]estimony describing the use of a dog to track an individual by scent, and demonstrating the accuracy of the track, does not involve seemingly infallible scientific devices, processes, or theories." Brooks, 950 P.2d at 652. In our view, the differences between a mechanical apparatus or standardized scientific procedure on the one hand, and a living, breathing, animate creature on the other, are weighty enough to take scent tracking outside the realm of processes ordinarily associated with the Frye standard. Although we acknowledge that Officer Nichols offered his thoughts on how bloodhounds might pick up scent, this was not the substantive thrust of his testimony. Instead, Nichols focused on Yogi's training, reliability, track record, and performance in the case at hand - all matters based on specialized knowledge he obtained as Yogi's handler. Contrary to Brooks' assertions, the reliability of scent tracking evidence is not dependant on the scientific explanation of canine olfaction.7

As we noted in Anderson, one reason for invoking the "general acceptance" standard found in Frye is the fear that jurors may give undue weight to evidence having an "aura of scientific infallibility." See Anderson, 637 P.2d at 361. However, unlike a scientific principle which a jury might assume to apply universally, the ability of a bloodhound to track is dependent upon precisely the sort of practical, experience-based matters that were at the heart of Nichols' testimony.

Further, we find that the "general acceptance" standard is cumbersome and of little value when applied beyond the realm of true "science." As some commentators have explained, Frye does not address the tough questions that arise on the cutting edge of science, as the test requires that the courts wait until science itself determines the validity of the scientific proposition in question. See Christopher B. Mueller & Laird C. Kirkpatrick, Federal Evidence 353 at 666 (2d ed. 1994)(observing that Frye "lets courts stand back from scientific controversy"). But this approach does not fit well when applied to expertise which, like that of the dog handler in this case, is based on years of experience and individualized "know-how" instead of some purportedly universal scientific principle. Indeed, even if there were a universally accepted theory explaining the canine ability to track scent, such consensus would be of little use in analyzing the evidentiary validity of the tracking efforts of a specific dog, working with a particular handler, in a particular case.

In light of these considerations, we conclude that canine scent tracking is not subject to the Frye test, and we decline to apply that standard here.


Brooks suggests that even if Frye does not apply to this case, we should adopt Daubert to exclude the evidence. We conclude that Daubert does not apply.

Daubert precluded further application of the Frye test in the federal courts based on the theory that, after adoption of the Federal Rules of Evidence, Fed. R. Evid. 702 displaced the "general acceptance" standard created more than fifty years earlier. See Daubert, 509 U.S. at 589. Although it held that Frye was no longer the appropriate test for novel science in the federal system, the Daubert opinion emphasized that Fed. R. Evid. 702 requires the courts to "ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Daubert, 509 U.S. at 589 (emphasis added). Focusing on the terms "scientific" and "knowledge" in Fed. R. Evid. 702, the court observed:

[I]n order to qualify as "scientific knowledge," an inference or assertion must be derived by the scientific method. Proposed testimony must be supported by appropriate validation - i.e., "good grounds," based on what is known. In short, the requirement that an expert's testimony pertain to "scientific knowledge" establishes a standard of evidentiary reliability.

Id. at 590.

In addition, the Daubert court reasoned that the particular considerations relevant to scientific reliability could be ascertained, and it undertook to outline several factors or "general considerations" to help guide the lower federal courts when making such determinations: (1) the testability of the scientific theory or technique; (2) whether the theory or technique had been subjected to peer review and publication; (3) the known or potential rate of error; (4) the existence or nonexistence of maintained standards; and (5) whether the theory or technique has general acceptance in a relevant scientific community. See id. Brooks suggests that these considerations are not satisfied in the instant case because, among other things, Officer Nichols could not offer a single testable theory on how dogs can track, and because neither Officer Nichols nor Yogi had been independently tested in a laboratory-style setting.

This court has neither explicitly endorsed nor rejected the Daubert analysis. In Lindsey, 892 P.2d at 288, we acknowledged the Daubert decision, but because we were construing the Colorado Rules of Evidence, we determined we were not bound by an interpretation of the federal rules as we are by constitutional analysis.8 We did not speak to the "relative merits of the Frye test or our corollary state rules of evidence," because the issue had not been fully briefed by the parties then before us. Id.

We are unwilling to adopt Daubert here and apply it for the first time to experience-based specialized knowledge which is not dependant on scientific explanation. The Daubert court itself limited its holding to the scientific realm. See Daubert, 509 U.S. at 590 n.8. In doing so, the court acknowledged that "[r]ule 702 also applies to 'technical, or other specialized knowledge,'" but limited its discussion "to the scientific context because that [was] the nature of the expertise" at issue in that case. Id. (emphasis added). Thus, the Daubert test is designed to analyze scientific evidence and not technical or other specialized knowledge. The factors listed by the court are grounded in the scientific process, and evaluate the quality of testing on an underlying scientific theory. Indeed, the Daubert court looked to external validation, as does the scientific process itself. However, like other nonscientific opinions, identification by scent tracking does not readily subject itself to external retesting. See Edward J. Imwinkelried, The Next Step After Daubert, 15 Cardozo L. Rev. 2271, 2283-84 (1994)(observing that "[n]either the essential test enunciated in Daubert, nor the factors listed by the Court are applicable to nonscientific opinion," in part because "nonscientific principles do not lend themselves to external retesting"). In order to apply Daubert, it would be necessary to reformulate the factors developed there. Accordingly, without commenting on whether we would adopt Daubert to analyze scientific knowledge,9 we will not apply factors designed to ascertain scientific validity to analyze the specialized knowledge before us.


Notwithstanding our conclusion that the standards from Frye and Daubert are inapplicable, nonscientific expertise remains subject to inquiry regarding validity and reliability. This inquiry is provided by the proper application of our rules of evidence. When proposed nonscientific expert testimony like the scent tracking evidence here draws an objection from the opposing party on the grounds that the subject matter is not sufficiently reliable to be presented to the jury, the common sense approach utilized in Campbell is useful. See Campbell, 814 P.2d at 7. That is, the trial court should first consider the standards for admissibility contained in CRE 702. If it finds that testimony on the subject indicated would be useful to the jury, and concludes the witness is qualified to opine on such matters, the court must then apply its discretionary authority under CRE 403. See id. at 8. Even though the proffered testimony might be admissible under CRE 702, a court may still consider whether the probative value is "substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence." CRE 403.

In this respect, the application of CRE 702 and CRE 403 are interrelated. A finding under CRE 403 that the nature of the proposed testimony is speculative or prejudicial, or that the link between the expertise and the hard evidence in the case is tenuous, necessarily weakens the likelihood that an opinion on the subject will be helpful to the trier of fact. Because the relevant factors applicable to the above inquiry will likely vary depending on the particular subject matter at hand, a trial court should make findings as it applies the CRE 702 and 403 analysis, tailoring its discussion to the foundational considerations relevant to the evidence before it.

We adopt the majority rule that other courts considering scent tracking have utilized to determine evidentiary reliability. Therefore, the elements of a proper foundation include: whether the dog is of a breed characterized by acute power of scent; whether the dog has been trained to follow a track by scent; whether the dog was found by experience to be reliable in pursuing human tracks; whether the dog was placed on the trail where the person being tracked was known to have been; and whether the tracking efforts took place within a reasonable time, given the abilities of the animal. See Terrell v. State, 239 A.2d 128 (Md. Ct. Spec. App. 1968) (comprehensive discussion of the history and development of the rule); see also State v. Roscoe, 700 P.2d 1312, 1321 (Ariz. 1985); Green v. State, 641 So.2d 391 (Fla. 1994); State v. Cole, 695 A.2d 1180, 1183 (Me. 1997; State v. Brewer, 875 S.W.2d 298 (Tenn. Crim. App. 1993). Further, most courts utilizing the majority rule exclude scent tracking evidence as too prejudicial where it is not corroborated by other independent evidence. See State v. Wainwright, 856 P.2d 163, 166 (Kan. Ct. App. 1993); People v. McPherson, 271 N.W.2d 228, 230 (Mich. Ct. App. 1978); State v. Loucks, 646 P.2d 480, 568 (Wash. 1983). We adopt this position as well. It is important to note that these foundational considerations are best utilized as a mechanism for conducting a proper CRE 702 and CRE 403 analysis, and not as a substitute for the general philosophy embodied in our rules of evidence. As such, the emphasis a court might wish to afford each of these points might vary depending on the facts of a particular case.

Here, although the first trial court judge erroneously concluded that this evidence should be analyzed under the Daubert factors, the second judge, who ultimately admitted the evidence, considered the admissibility standards contained in the rules of evidence and the considerations traditionally examined by courts considering scent tracking.10 In doing so, the court found that Yogi was in fact a purebred bloodhound trained and found by experience to have the ability to track, and had been introduced to the scent from footprints made by the suspect a very short time after the trail was laid. The court also thought it significant that there was substantial evidence corroborating Yogi's find, including the fact that officers in hot pursuit followed the suspect's footprints to the same location, and the discovery of burglary tools and items taken from the victim's home on or about Brooks' person. In light of the extensive corroboration, the trial court found that Officer Nichols' testimony was not unduly prejudicial.

Based on our review of the record before us, including the evidence corroborating Yogi's find, we conclude that the second trial court judge did not abuse his discretion in allowing Officer Nichols' testimony, even in light of the earlier court's ruling that the evidence would only be admissible, if at all, under CRE 701.11 See Huntoon, 969 P.2d at 690 (court's decision to allow expert testimony will not be disturbed absent abuse of discretion); Williams, 790 P.2d 796, 798 (Colo. 1990).


In sum, we hold that testimony pertaining to scent tracking by a trained police dog is not readily subjected to standards created with scientific evidence in mind. Accordingly, the conventional safeguards found in CRE 702 and CRE 403 govern the admissibility of such testimony. We further set forth the elements of a proper foundation for the admission of scent tracking evidence, including a corroboration requirement, but emphasize that the focus of the inquiry here is embodied in CRE 702 and CRE 403. Because the information presented on the scent tracking in this case was properly before the jury, we affirm the judgment of the court of appeals.



1 We granted certiorari on the following questions:

Is dog tracking evidence "scientific" evidence the admissibility of which is governed by the test set forth in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), or Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

Did the court of appeals err in finding that the dog handler's testimony provided sufficient foundation under CRE 702 for the admission of evidence that the dog had identified the defendant as the perpetrator of the crimes charged.

2 See § 8-4-203, 6 C.R.S. (1998).

3 See §§ 18-2-101 and 18-4-401, 6 C.R.S. (1998).

4 See § 18-4-205, 6 C.R.S. (1998).

5 On one occasion, Officer Nichols and Yogi were unable to locate the other side of one of these fences due to the layout of the neighborhood and the cul-de-sacs in the area. When this happened, Nichols and Yogi were driven to another location where they could more easily access and continue to follow the trail.

6 We have serious doubts as to whether scent tracking evidence even approaches the sort of novelty that is ordinarily associated with a Frye analysis. The use of scent tracking enjoys longstanding and colorful antecedents. Indeed, courts have recognized that some dogs possess the ability to track by scent even in very early decisions appearing in the American case reporters. See, e.g., Hodge v. State, 13 So. 385 (Ala. 1893); Davis v. State, 35 So. 76 (Fla. 1903); Pedigo v. Commonwealth, 44 S.W. 143 (Ky. 1898). Brooks does not forcefully argue that canine scenting is itself "novel," but suggests certain methods used in his case, such as "drop trailing" might fit within this realm. Because our resolution of this issue does not depend on the novelty of the evidence at hand, we need not address this contention.

7 Brooks' arguments suggest that whenever one can find "science" by scratching beneath the surface of expert testimony, the validation rules governing scientific evidence would have to apply. We reject this view. Nearly any topic that might be the subject of courtroom testimony can be characterized in such a way as to expose scientific principles. This is simply inescapable because the human experience takes place in the physical world - a world we explain with the laws of physics, chemistry, mathematics, and the like. Even a common eyewitness identification by way of lay testimony could involve "science" if one were to examine the scientific principles that underlie human vision and perception. The fact that some aspect of a witness' testimony can be described in scientific terms does not mean, ipso facto, that the jury must understand the science in order to find the testimony helpful. Instead, the courts should determine whether the opinion being offered either depends on scientific axioms, or is based on scientific theory, analysis, or experimentation.

8 We note that the primary rationale stated by the Daubert court for its rejection of Frye in the federal sphere is inapplicable when applied to this state. Although CRE 702 is worded exactly as its federal counterpart, our adoption of rule 702 could not have displaced Frye. Contrary to the experience of the federal courts, the Frye test was first applied in Colorado after the adoption of rule 702. See Anderson, 637 P.2d at 358 (applying Frye subsequent to adoption of Colorado Rules of Evidence); see also Fishback, 851 P.2d at 890 n.15 (noting that CRE 702 did not displace Frye in this state).

9 We are of the view that the Daubert standards do not assist the analysis of the "specialized knowledge" evidence at issue here. This is true even in light of the fact that Daubert is potentially applicable to a broader category of evidence than that covered by Frye. See Daubert, 509 U.S. at 592 n.11 (explaining that its reasoning is applicable beyond the realm of novel or unconventional science); see also Christopher B. Mueller and Laird C. Kirkpatrick, Federal Evidence § 353 at 667 (2d ed. 1994)(suggesting it might be wrong in a Daubert jurisdiction to exempt social science as some courts had done with Frye). Notably, the United States Supreme Court is currently deciding whether the Daubert factors should be used to analyze evidence that presents a mix of applied scientific knowledge and knowledge gained through experience. See Kumho Tire Co. v. Carmichael, 118 S. Ct. 2339 (1998)(granting certiorari to decide whether a trial judge may consider the factors set out in Daubert, in a Rule 702 analysis of the admissibility of an engineering expert's testimony). To the extent that the Supreme Court might alter the Daubert standard in Kumho, we are without the benefit of such analysis here.

10 Brooks asserts that review here is limited to whether this evidence was proper under CRE 701, the rule under which it was admitted. We disagree. The admission of evidence can be defended on any theory of admissibility, even one not utilized by the trial court. See Lindsey, 892 P.2d at 292; People v. Jenkins, 768 P.2d 727, 730 (Colo. App. 1988)(refusing to reverse conviction where trial court reached correct result by applying incorrect evidentiary analysis) People v. Mathes, 703 P.2d 608, 610 (Colo. App. 1985)(admissible evidence does not become inadmissible where trial court relies on wrong rule).

11 Brooks complains that the decision, on the eve of trial, by the second trial court to admit the evidence was prejudicial because he was unable to obtain an expert so late in the proceedings. However, our review of the record indicates that the first trial court judge clearly left open the possibility of testimony on scent tracking being introduced at the discretion of the second judge. Accordingly, Brooks had notice that such evidence might be admitted in some form, and ample opportunity to proffer an expert to meet such evidence.

These opinions are not final. They may be modified, changed or withdrawn in accordance with Rules 40 and 49 of the Colorado Appellate Rules. Changes to or modifications of these opinions resulting from any action taken by the Court of Appeals or the Supreme Court are not incorporated here.